fix: only fall back to polling when streaming is explicitly False

Addresses review: agents that leave streaming unset (None) should get the default StreamingHandler, not polling. Only agents that explicitly set streaming=False trigger the polling fallback.
fix: respect agent card streaming capability in A2A delegation
2026-04-15 23:42:37 +00:00 · 2026-04-12 16:38:44 -07:00 · 2026-04-12 16:23:27 -07:00
41 changed files with 685 additions and 1236 deletions
--- a/docs/ar/changelog.mdx
+++ b/docs/ar/changelog.mdx
@@ -4,97 +4,6 @@ description: "تحديثات المنتج والتحسينات وإصلاحات
 icon: "clock"
 mode: "wide"
 ---
-<Update label="16 أبريل 2026">
-  ## v1.14.2rc1
-
-  [عرض الإصدار على GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2rc1)
-
-  ## ما الذي تغير
-
-  ### إصلاحات الأخطاء
-  - إصلاح معالجة مخططات JSON الدائرية في أداة MCP
-  - إصلاح ثغرة أمنية من خلال تحديث python-multipart إلى 0.0.26
-  - إصلاح ثغرة أمنية من خلال تحديث pypdf إلى 6.10.1
-
-  ### الوثائق
-  - تحديث سجل التغييرات والإصدار لـ v1.14.2a5
-
-  ## المساهمون
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="15 أبريل 2026">
-  ## v1.14.2a5
-
-  [عرض الإصدار على GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a5)
-
-  ## ما الذي تغير
-
-  ### الوثائق
-  - تحديث سجل التغييرات والإصدار لـ v1.14.2a4
-
-  ## المساهمون
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="15 أبريل 2026">
-  ## v1.14.2a4
-
-  [عرض الإصدار على GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a4)
-
-  ## ما الذي تغير
-
-  ### الميزات
-  - إضافة تلميحات استئناف إلى إصدار أدوات المطورين عند الفشل
-
-  ### إصلاحات الأخطاء
-  - إصلاح توجيه وضع الصرامة إلى واجهة برمجة تطبيقات Bedrock Converse
-  - إصلاح إصدار pytest إلى 9.0.3 لثغرة الأمان GHSA-6w46-j5rx-g56g
-  - رفع الحد الأدنى لـ OpenAI إلى >=2.0.0
-
-  ### الوثائق
-  - تحديث سجل التغييرات والإصدار لـ v1.14.2a3
-
-  ## المساهمون
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="13 أبريل 2026">
-  ## v1.14.2a3
-
-  [عرض الإصدار على GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a3)
-
-  ## ما الذي تغير
-
-  ### الميزات
-  - إضافة واجهة سطر الأوامر للتحقق من النشر
-  - تحسين سهولة استخدام تهيئة LLM
-
-  ### إصلاحات الأخطاء
-  - تجاوز pypdf و uv إلى إصدارات مصححة لـ CVE-2026-40260 و GHSA-pjjw-68hj-v9mw
-  - ترقية requests إلى >=2.33.0 لمعالجة ثغرة ملف مؤقت CVE
-  - الحفاظ على معلمات استدعاء أداة Bedrock من خلال إزالة القيمة الافتراضية الصحيحة
-  - تنظيف مخططات الأدوات لوضع صارم
-  - إصلاح اختبار تسلسل تضمين MemoryRecord
-
-  ### الوثائق
-  - تنظيف لغة A2A الخاصة بالمؤسسات
-  - إضافة وثائق ميزات A2A الخاصة بالمؤسسات
-  - تحديث وثائق A2A الخاصة بالمصادر المفتوحة
-  - تحديث سجل التغييرات والإصدار لـ v1.14.2a2
-
-  ## المساهمون
-
-  @Yanhu007, @greysonlalonde
-
-</Update>
-
 <Update label="10 أبريل 2026">
  ## v1.14.2a2

--- a/docs/docs.json
+++ b/docs/docs.json
@@ -392,8 +392,7 @@
                      "en/enterprise/features/marketplace",
                      "en/enterprise/features/agent-repositories",
                      "en/enterprise/features/tools-and-integrations",
-                      "en/enterprise/features/pii-trace-redactions",
-                      "en/enterprise/features/a2a"
+                      "en/enterprise/features/pii-trace-redactions"
                    ]
                  },
                  {
@@ -866,8 +865,7 @@
                      "en/enterprise/features/marketplace",
                      "en/enterprise/features/agent-repositories",
                      "en/enterprise/features/tools-and-integrations",
-                      "en/enterprise/features/pii-trace-redactions",
-                      "en/enterprise/features/a2a"
+                      "en/enterprise/features/pii-trace-redactions"
                    ]
                  },
                  {
@@ -1340,8 +1338,7 @@
                      "en/enterprise/features/marketplace",
                      "en/enterprise/features/agent-repositories",
                      "en/enterprise/features/tools-and-integrations",
-                      "en/enterprise/features/pii-trace-redactions",
-                      "en/enterprise/features/a2a"
+                      "en/enterprise/features/pii-trace-redactions"
                    ]
                  },
                  {
@@ -1814,8 +1811,7 @@
                      "en/enterprise/features/marketplace",
                      "en/enterprise/features/agent-repositories",
                      "en/enterprise/features/tools-and-integrations",
-                      "en/enterprise/features/pii-trace-redactions",
-                      "en/enterprise/features/a2a"
+                      "en/enterprise/features/pii-trace-redactions"
                    ]
                  },
                  {
@@ -2287,8 +2283,7 @@
                      "en/enterprise/features/marketplace",
                      "en/enterprise/features/agent-repositories",
                      "en/enterprise/features/tools-and-integrations",
-                      "en/enterprise/features/pii-trace-redactions",
-                      "en/enterprise/features/a2a"
+                      "en/enterprise/features/pii-trace-redactions"
                    ]
                  },
                  {
@@ -2759,8 +2754,7 @@
                      "en/enterprise/features/marketplace",
                      "en/enterprise/features/agent-repositories",
                      "en/enterprise/features/tools-and-integrations",
-                      "en/enterprise/features/pii-trace-redactions",
-                      "en/enterprise/features/a2a"
+                      "en/enterprise/features/pii-trace-redactions"
                    ]
                  },
                  {
@@ -3231,8 +3225,7 @@
                      "en/enterprise/features/marketplace",
                      "en/enterprise/features/agent-repositories",
                      "en/enterprise/features/tools-and-integrations",
-                      "en/enterprise/features/pii-trace-redactions",
-                      "en/enterprise/features/a2a"
+                      "en/enterprise/features/pii-trace-redactions"
                    ]
                  },
                  {
@@ -3705,8 +3698,7 @@
                      "en/enterprise/features/marketplace",
                      "en/enterprise/features/agent-repositories",
                      "en/enterprise/features/tools-and-integrations",
-                      "en/enterprise/features/pii-trace-redactions",
-                      "en/enterprise/features/a2a"
+                      "en/enterprise/features/pii-trace-redactions"
                    ]
                  },
                  {
@@ -4177,8 +4169,7 @@
                      "en/enterprise/features/marketplace",
                      "en/enterprise/features/agent-repositories",
                      "en/enterprise/features/tools-and-integrations",
-                      "en/enterprise/features/pii-trace-redactions",
-                      "en/enterprise/features/a2a"
+                      "en/enterprise/features/pii-trace-redactions"
                    ]
                  },
                  {
@@ -4652,8 +4643,7 @@
                      "en/enterprise/features/marketplace",
                      "en/enterprise/features/agent-repositories",
                      "en/enterprise/features/tools-and-integrations",
-                      "en/enterprise/features/pii-trace-redactions",
-                      "en/enterprise/features/a2a"
+                      "en/enterprise/features/pii-trace-redactions"
                    ]
                  },
                  {
--- a/docs/en/changelog.mdx
+++ b/docs/en/changelog.mdx
@@ -4,97 +4,6 @@ description: "Product updates, improvements, and bug fixes for CrewAI"
 icon: "clock"
 mode: "wide"
 ---
-<Update label="Apr 16, 2026">
-  ## v1.14.2rc1
-
-  [View release on GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2rc1)
-
-  ## What's Changed
-
-  ### Bug Fixes
-  - Fix handling of cyclic JSON schemas in MCP tool resolution
-  - Fix vulnerability by bumping python-multipart to 0.0.26
-  - Fix vulnerability by bumping pypdf to 6.10.1
-
-  ### Documentation
-  - Update changelog and version for v1.14.2a5
-
-  ## Contributors
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="Apr 15, 2026">
-  ## v1.14.2a5
-
-  [View release on GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a5)
-
-  ## What's Changed
-
-  ### Documentation
-  - Update changelog and version for v1.14.2a4
-
-  ## Contributors
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="Apr 15, 2026">
-  ## v1.14.2a4
-
-  [View release on GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a4)
-
-  ## What's Changed
-
-  ### Features
-  - Add resume hints to devtools release on failure
-
-  ### Bug Fixes
-  - Fix strict mode forwarding to Bedrock Converse API
-  - Fix pytest version to 9.0.3 for security vulnerability GHSA-6w46-j5rx-g56g
-  - Bump OpenAI lower bound to >=2.0.0
-
-  ### Documentation
-  - Update changelog and version for v1.14.2a3
-
-  ## Contributors
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="Apr 13, 2026">
-  ## v1.14.2a3
-
-  [View release on GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a3)
-
-  ## What's Changed
-
-  ### Features
-  - Add deploy validation CLI
-  - Improve LLM initialization ergonomics
-
-  ### Bug Fixes
-  - Override pypdf and uv to patched versions for CVE-2026-40260 and GHSA-pjjw-68hj-v9mw
-  - Upgrade requests to >=2.33.0 for CVE temp file vulnerability
-  - Preserve Bedrock tool call arguments by removing truthy default
-  - Sanitize tool schemas for strict mode
-  - Deflake MemoryRecord embedding serialization test
-
-  ### Documentation
-  - Clean up enterprise A2A language
-  - Add enterprise A2A feature documentation
-  - Update OSS A2A documentation
-  - Update changelog and version for v1.14.2a2
-
-  ## Contributors
-
-  @Yanhu007, @greysonlalonde
-
-</Update>
-
 <Update label="Apr 10, 2026">
  ## v1.14.2a2

--- a/docs/en/enterprise/features/a2a.mdx
+++ b/docs/en/enterprise/features/a2a.mdx
@@ -1,227 +0,0 @@
---
-title: A2A on AMP
-description: Production-grade Agent-to-Agent communication with distributed state and multi-scheme authentication
-icon: "network-wired"
-mode: "wide"
---
-
-<Warning>
-A2A server agents on AMP are in early release. APIs may change in future versions.
-</Warning>
-
-## Overview
-
-CrewAI AMP extends the open-source [A2A protocol implementation](/en/learn/a2a-agent-delegation) with production infrastructure for deploying distributed agents at scale. AMP supports A2A protocol versions 0.2 and 0.3. When you deploy a crew or agent with A2A server configuration to AMP, the platform automatically provisions distributed state management, authentication, multi-transport endpoints, and lifecycle management.
-
-<Note>
-  For A2A protocol fundamentals, client/server configuration, and authentication schemes, see the [A2A Agent Delegation](/en/learn/a2a-agent-delegation) documentation. This page covers what AMP adds on top of the open-source implementation.
-</Note>
-
-### Usage
-
-Add `A2AServerConfig` to any agent in your crew and deploy to AMP. The platform detects agents with server configuration and automatically registers A2A endpoints, generates agent cards, and provisions the infrastructure described below.
-
-```python
-from crewai import Agent, Crew, Task
-from crewai.a2a import A2AServerConfig
-from crewai.a2a.auth import EnterpriseTokenAuth
-
-agent = Agent(
-    role="Data Analyst",
-    goal="Analyze datasets and provide insights",
-    backstory="Expert data scientist with statistical analysis skills",
-    llm="gpt-4o",
-    a2a=A2AServerConfig(
-        auth=EnterpriseTokenAuth()
-    )
-)
-
-task = Task(
-    description="Analyze the provided dataset",
-    expected_output="Statistical summary with key insights",
-    agent=agent
-)
-
-crew = Crew(agents=[agent], tasks=[task])
-```
-
-After [deploying to AMP](/en/enterprise/guides/deploy-to-amp), the platform registers two levels of A2A endpoints:
-
- **Crew-level**: an aggregate agent card at `/.well-known/agent-card.json` where each agent with `A2AServerConfig` is listed as a skill, with a JSON-RPC endpoint at `/a2a`
- **Per-agent**: isolated agent cards and JSON-RPC endpoints mounted at `/a2a/agents/{role}/`, each with its own tenancy
-
-Clients can interact with the crew as a whole or target a specific agent directly. To route a request to a specific agent through the crew-level endpoint, include `"target_agent"` in the message metadata with the agent's slugified role name (e.g., `"data-analyst"` for an agent with role `"Data Analyst"`). If no `target_agent` is provided, the request is handled by the first agent in the crew.
-
-See [A2A Agent Delegation](/en/learn/a2a-agent-delegation#server-configuration-options) for the full list of `A2AServerConfig` options.
-
-<Warning>
-  Per the A2A protocol, agent cards are publicly accessible to enable discovery. This includes both the crew-level card at `/.well-known/agent-card.json` and per-agent cards at `/a2a/agents/{role}/.well-known/agent-card.json`. Do not include sensitive information in agent names, descriptions, or skill definitions.
-</Warning>
-
-### File Inputs and Structured Output
-
-A2A on AMP supports passing files and requesting structured output in both directions. Clients can send files as `FilePart`s and request structured responses by embedding a JSON schema in the message. Server agents receive files as `input_files` on the task, and return structured data as `DataPart`s when a schema is provided. See [File Inputs and Structured Output](/en/learn/a2a-agent-delegation#file-inputs-and-structured-output) for details.
-
-### What AMP Adds
-
-<CardGroup cols={2}>
-  <Card title="Distributed State" icon="database">
-    Persistent task, context, and result storage
-  </Card>
-  <Card title="Enterprise Authentication" icon="shield-halved">
-    OIDC, OAuth2, mTLS, and Enterprise token validation beyond simple bearer tokens
-  </Card>
-  <Card title="gRPC Transport" icon="bolt">
-    Full gRPC server with TLS and authentication
-  </Card>
-  <Card title="Context Lifecycle" icon="clock-rotate-left">
-    Automatic idle detection, expiration, and cleanup of long-running conversations
-  </Card>
-  <Card title="Signed Webhooks" icon="signature">
-    HMAC-SHA256 signed push notifications with replay protection
-  </Card>
-  <Card title="Multi-Transport" icon="arrows-split-up-and-left">
-    REST, JSON-RPC, and gRPC endpoints served simultaneously from a single deployment
-  </Card>
-</CardGroup>
-
---
-
-## Distributed State Management
-
-In the open-source implementation, task and context state lives in memory on a single process. AMP replaces this with persistent, distributed stores.
-
-### Storage Layers
-
-| Store | Purpose |
-|---|---|
-| **Task Store** | Persists A2A task state and metadata |
-| **Context Store** | Tracks conversation context, creation time, last activity, and associated tasks |
-| **Result Store** | Caches task results for retrieval |
-| **Push Config Store** | Manages webhook subscriptions per task |
-
-Multiple A2A deployments are automatically isolated from each other, preventing data collisions when sharing infrastructure.
-
---
-
-## Enterprise Authentication
-
-AMP supports six authentication schemes for incoming A2A requests, configurable per deployment. Authentication works across both HTTP and gRPC transports.
-
-| Scheme | Description | Use Case |
-|---|---|---|
-| **SimpleTokenAuth** | Static bearer token from `AUTH_TOKEN` env var | Development, simple deployments |
-| **EnterpriseTokenAuth** | Token verification via CrewAI PlusAPI with integration token claims | AMP-to-AMP agent communication |
-| **OIDCAuth** | OpenID Connect JWT validation with JWKS endpoint caching | Enterprise SSO integration |
-| **OAuth2ServerAuth** | OAuth2 with configurable scopes | Fine-grained access control |
-| **APIKeyServerAuth** | API key validation via header or query parameter | Third-party integrations |
-| **MTLSServerAuth** | Mutual TLS certificate-based authentication | Zero-trust environments |
-
-The configured auth scheme automatically populates the agent card's `securitySchemes` and `security` fields. Clients discover authentication requirements by fetching the agent card before making requests.
-
---
-
-## Extended Agent Cards
-
-AMP supports role-based skill visibility through extended agent cards. Unauthenticated users see the standard agent card with public skills. Authenticated users receive an extended card with additional capabilities.
-
-This enables patterns like:
- Public agents that expose basic skills to anyone, with advanced skills available to authenticated clients
- Internal agents that advertise different capabilities based on the caller's identity
-
---
-
-## gRPC Transport
-
-If enabled, AMP provides full gRPC support alongside the default JSON-RPC transport.
-
- **TLS termination** with configurable certificate and key paths
- **gRPC reflection** for debugging with tools like `grpcurl`
- **Authentication** using the same schemes available for HTTP
- **Extension validation** ensuring clients support required protocol extensions
- **Version negotiation** across A2A protocol versions 0.2 and 0.3
-
-For deployments exposing multiple agents, AMP automatically allocates per-agent gRPC ports and coordinates TLS, startup, and shutdown across all servers.
-
---
-
-## Context Lifecycle Management
-
-AMP tracks the lifecycle of A2A conversation contexts and automatically manages cleanup.
-
-### Lifecycle States
-
-| State | Condition | Action |
-|---|---|---|
-| **Active** | Context has recent activity | None |
-| **Idle** | No activity for a configured period | Marked idle, event emitted |
-| **Expired** | Context exceeds its maximum lifetime | Marked expired, associated tasks cleaned up, event emitted |
-
-A background cleanup task runs hourly to scan for idle and expired contexts. All state transitions emit CrewAI events that integrate with the platform's observability features.
-
---
-
-## Signed Push Notifications
-
-When an A2A agent sends push notifications to a client webhook, AMP signs each request with HMAC-SHA256 to ensure integrity and prevent tampering.
-
-### Signature Headers
-
-| Header | Purpose |
-|---|---|
-| `X-A2A-Signature` | HMAC-SHA256 signature in `sha256={hex_digest}` format |
-| `X-A2A-Signature-Timestamp` | Unix timestamp bound to the signature |
-| `X-A2A-Notification-Token` | Optional notification auth token |
-
-### Security Properties
-
- **Integrity**: payload cannot be modified without invalidating the signature
- **Replay protection**: signatures are timestamp-bound with a configurable tolerance window
- **Retry with backoff**: failed deliveries retry with exponential backoff
-
---
-
-## Distributed Event Streaming
-
-In the open-source implementation, SSE streaming works within a single process. AMP propagates SSE events across instances so that clients receive updates even when the instance holding the streaming connection differs from the instance executing the task.
-
---
-
-## Multi-Transport Endpoints
-
-AMP serves REST and JSON-RPC by default. gRPC is available as an additional transport if enabled.
-
-| Transport | Path Convention | Description |
-|---|---|---|
-| **REST** | `/v1/message:send`, `/v1/message:stream`, `/v1/tasks` | Google API conventions |
-| **JSON-RPC** | Standard A2A JSON-RPC endpoint | Default A2A protocol transport |
-| **gRPC** | Per-agent port allocation | Optional, high-performance binary protocol |
-
-All active transports share the same authentication, version negotiation, and extension validation. Agent cards are generated from agent and crew metadata — roles, goals, and tools become skills and descriptions — and automatically include interfaces for each active transport. They can also be manually configured via `A2AServerConfig`.
-
---
-
-## Version and Extension Negotiation
-
-AMP validates A2A protocol versions and extensions at the transport layer.
-
-### Version Negotiation
-
- Clients send the `A2A-Version` header with their preferred version
- AMP validates against supported versions (0.2, 0.3) and falls back to 0.3 if unspecified
- The negotiated version is returned in the response headers
-
-### Extension Validation
-
- Clients declare supported extensions via the `X-A2A-Extensions` header
- AMP validates that clients support all extensions the agent requires
- Requests from clients missing required extensions receive an `UnsupportedExtensionError`
-
---
-
-## Next Steps
-
- [A2A Agent Delegation](/en/learn/a2a-agent-delegation) — A2A protocol fundamentals and configuration
- [A2UI](/en/learn/a2ui) — Interactive UI rendering over A2A
- [Deploy to AMP](/en/enterprise/guides/deploy-to-amp) — General deployment guide
- [Webhook Streaming](/en/enterprise/features/webhook-streaming) — Event streaming for deployed automations
--- a/docs/en/learn/a2a-agent-delegation.mdx
+++ b/docs/en/learn/a2a-agent-delegation.mdx
@@ -7,10 +7,6 @@ mode: "wide"

 ## A2A Agent Delegation

-<Info>
-  Deploying A2A agents to production? See [A2A on AMP](/en/enterprise/features/a2a) for distributed state, enterprise authentication, gRPC transport, and horizontal scaling.
-</Info>
-
 CrewAI treats [A2A protocol](https://a2a-protocol.org/latest/) as a first-class delegation primitive, enabling agents to delegate tasks, request information, and collaborate with remote agents, as well as act as A2A-compliant server agents.
 In client mode, agents autonomously choose between local execution and remote delegation based on task requirements.

@@ -100,28 +96,24 @@ The `A2AClientConfig` class accepts the following parameters:
  Update mechanism for receiving task status. Options: `StreamingConfig`, `PollingConfig`, or `PushNotificationConfig`.
 </ParamField>

+<ParamField path="transport_protocol" type="Literal['JSONRPC', 'GRPC', 'HTTP+JSON']" default="JSONRPC">
+  Transport protocol for A2A communication. Options: `JSONRPC` (default), `GRPC`, or `HTTP+JSON`.
+</ParamField>
+
 <ParamField path="accepted_output_modes" type="list[str]" default='["application/json"]'>
  Media types the client can accept in responses.
 </ParamField>

+<ParamField path="supported_transports" type="list[str]" default='["JSONRPC"]'>
+  Ordered list of transport protocols the client supports.
+</ParamField>
+
+<ParamField path="use_client_preference" type="bool" default="False">
+  Whether to prioritize client transport preferences over server.
+</ParamField>
+
 <ParamField path="extensions" type="list[str]" default="[]">
-  A2A protocol extension URIs the client supports.
-</ParamField>
-
-<ParamField path="client_extensions" type="list[A2AExtension]" default="[]">
-  Client-side processing hooks for tool injection, prompt augmentation, and response modification.
-</ParamField>
-
-<ParamField path="transport" type="ClientTransportConfig" default="ClientTransportConfig()">
-  Transport configuration including preferred transport, supported transports for negotiation, and protocol-specific settings (gRPC message sizes, keepalive, etc.).
-</ParamField>
-
-<ParamField path="transport_protocol" type="Literal['JSONRPC', 'GRPC', 'HTTP+JSON']" default="None">
-  **Deprecated**: Use `transport=ClientTransportConfig(preferred=...)` instead.
-</ParamField>
-
-<ParamField path="supported_transports" type="list[str]" default="None">
-  **Deprecated**: Use `transport=ClientTransportConfig(supported=...)` instead.
+  Extension URIs the client supports.
 </ParamField>

 ## Authentication
@@ -413,7 +405,11 @@ agent = Agent(
  Preferred endpoint URL. If set, overrides the URL passed to `to_agent_card()`.
 </ParamField>

-<ParamField path="protocol_version" type="str" default="0.3.0">
+<ParamField path="preferred_transport" type="Literal['JSONRPC', 'GRPC', 'HTTP+JSON']" default="JSONRPC">
+  Transport protocol for the preferred endpoint.
+</ParamField>
+
+<ParamField path="protocol_version" type="str" default="0.3">
  A2A protocol version this agent supports.
 </ParamField>

@@ -445,36 +441,8 @@ agent = Agent(
  Whether agent provides extended card to authenticated users.
 </ParamField>

-<ParamField path="extended_skills" type="list[AgentSkill]" default="[]">
-  Additional skills visible only to authenticated users in the extended agent card.
-</ParamField>
-
-<ParamField path="signing_config" type="AgentCardSigningConfig" default="None">
-  Configuration for signing the AgentCard with JWS. Supports RS256, ES256, PS256, and related algorithms.
-</ParamField>
-
-<ParamField path="server_extensions" type="list[ServerExtension]" default="[]">
-  Server-side A2A protocol extensions with `on_request`/`on_response` hooks that modify agent behavior.
-</ParamField>
-
-<ParamField path="push_notifications" type="ServerPushNotificationConfig" default="None">
-  Configuration for outgoing push notifications, including HMAC-SHA256 signing secret.
-</ParamField>
-
-<ParamField path="transport" type="ServerTransportConfig" default="ServerTransportConfig()">
-  Transport configuration including preferred transport, gRPC server settings, JSON-RPC paths, and HTTP+JSON settings.
-</ParamField>
-
-<ParamField path="auth" type="ServerAuthScheme" default="None">
-  Authentication scheme for incoming A2A requests. Defaults to `SimpleTokenAuth` using the `AUTH_TOKEN` environment variable.
-</ParamField>
-
-<ParamField path="preferred_transport" type="Literal['JSONRPC', 'GRPC', 'HTTP+JSON']" default="None">
-  **Deprecated**: Use `transport=ServerTransportConfig(preferred=...)` instead.
-</ParamField>
-
-<ParamField path="signatures" type="list[AgentCardSignature]" default="None">
-  **Deprecated**: Use `signing_config=AgentCardSigningConfig(...)` instead.
+<ParamField path="signatures" type="list[AgentCardSignature]" default="[]">
+  JSON Web Signatures for the AgentCard.
 </ParamField>

 ### Combined Client and Server
@@ -500,14 +468,6 @@ agent = Agent(
 )
 ```

-### File Inputs and Structured Output
-
-A2A supports passing files and requesting structured output in both directions.
-
-**Client side**: When delegating to a remote A2A agent, files from the task's `input_files` are sent as `FilePart`s in the outgoing message. If `response_model` is set on the `A2AClientConfig`, the Pydantic model's JSON schema is embedded in the message metadata, requesting structured output from the remote agent.
-
-**Server side**: Incoming `FilePart`s are extracted and passed to the agent's task as `input_files`. If the client included a JSON schema, the server creates a response model from it and applies it to the task. When the agent returns structured data, the response is sent back as a `DataPart` rather than plain text.
-
 ## Best Practices

 <CardGroup cols={2}>
--- a/docs/ko/changelog.mdx
+++ b/docs/ko/changelog.mdx
@@ -4,97 +4,6 @@ description: "CrewAI의 제품 업데이트, 개선 사항 및 버그 수정"
 icon: "clock"
 mode: "wide"
 ---
-<Update label="2026년 4월 16일">
-  ## v1.14.2rc1
-
-  [GitHub 릴리스 보기](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2rc1)
-
-  ## 변경 사항
-
-  ### 버그 수정
-  - MCP 도구 해상도에서 순환 JSON 스키마 처리 수정
-  - python-multipart를 0.0.26으로 업데이트하여 취약점 수정
-  - pypdf를 6.10.1로 업데이트하여 취약점 수정
-
-  ### 문서
-  - v1.14.2a5에 대한 변경 로그 및 버전 업데이트
-
-  ## 기여자
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="2026년 4월 15일">
-  ## v1.14.2a5
-
-  [GitHub 릴리스 보기](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a5)
-
-  ## 변경 사항
-
-  ### 문서
-  - v1.14.2a4의 변경 로그 및 버전 업데이트
-
-  ## 기여자
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="2026년 4월 15일">
-  ## v1.14.2a4
-
-  [GitHub 릴리스 보기](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a4)
-
-  ## 변경 사항
-
-  ### 기능
-  - 실패 시 devtools 릴리스에 이력서 힌트 추가
-
-  ### 버그 수정
-  - Bedrock Converse API로의 엄격 모드 포워딩 수정
-  - 보안 취약점 GHSA-6w46-j5rx-g56g에 대해 pytest 버전을 9.0.3으로 수정
-  - OpenAI 하한을 >=2.0.0으로 상향 조정
-
-  ### 문서
-  - v1.14.2a3에 대한 변경 로그 및 버전 업데이트
-
-  ## 기여자
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="2026년 4월 13일">
-  ## v1.14.2a3
-
-  [GitHub 릴리스 보기](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a3)
-
-  ## 변경 사항
-
-  ### 기능
-  - 배포 검증 CLI 추가
-  - LLM 초기화 사용성 개선
-
-  ### 버그 수정
-  - CVE-2026-40260 및 GHSA-pjjw-68hj-v9mw에 대한 패치된 버전으로 pypdf 및 uv 재정의
-  - CVE 임시 파일 취약점에 대해 requests를 >=2.33.0으로 업그레이드
-  - 진리값 기본값을 제거하여 Bedrock 도구 호출 인수 보존
-  - 엄격 모드를 위한 도구 스키마 정리
-  - MemoryRecord 임베딩 직렬화 테스트의 불안정성 제거
-
-  ### 문서
-  - 기업 A2A 언어 정리
-  - 기업 A2A 기능 문서 추가
-  - OSS A2A 문서 업데이트
-  - v1.14.2a2에 대한 변경 로그 및 버전 업데이트
-
-  ## 기여자
-
-  @Yanhu007, @greysonlalonde
-
-</Update>
-
 <Update label="2026년 4월 10일">
  ## v1.14.2a2

--- a/docs/pt-BR/changelog.mdx
+++ b/docs/pt-BR/changelog.mdx
@@ -4,97 +4,6 @@ description: "Atualizações de produto, melhorias e correções do CrewAI"
 icon: "clock"
 mode: "wide"
 ---
-<Update label="16 abr 2026">
-  ## v1.14.2rc1
-
-  [Ver release no GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2rc1)
-
-  ## O que Mudou
-
-  ### Correções de Bugs
-  - Corrigir o manuseio de esquemas JSON cíclicos na resolução da ferramenta MCP
-  - Corrigir vulnerabilidade atualizando python-multipart para 0.0.26
-  - Corrigir vulnerabilidade atualizando pypdf para 6.10.1
-
-  ### Documentação
-  - Atualizar o changelog e a versão para v1.14.2a5
-
-  ## Contribuidores
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="15 abr 2026">
-  ## v1.14.2a5
-
-  [Ver release no GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a5)
-
-  ## O que Mudou
-
-  ### Documentação
-  - Atualizar changelog e versão para v1.14.2a4
-
-  ## Contribuidores
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="15 abr 2026">
-  ## v1.14.2a4
-
-  [Ver release no GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a4)
-
-  ## O que Mudou
-
-  ### Recursos
-  - Adicionar dicas de retomar ao release do devtools em caso de falha
-
-  ### Correções de Bugs
-  - Corrigir o encaminhamento do modo estrito para a API Bedrock Converse
-  - Corrigir a versão do pytest para 9.0.3 devido à vulnerabilidade de segurança GHSA-6w46-j5rx-g56g
-  - Aumentar o limite inferior do OpenAI para >=2.0.0
-
-  ### Documentação
-  - Atualizar o changelog e a versão para v1.14.2a3
-
-  ## Contribuidores
-
-  @greysonlalonde
-
-</Update>
-
-<Update label="13 abr 2026">
-  ## v1.14.2a3
-
-  [Ver release no GitHub](https://github.com/crewAIInc/crewAI/releases/tag/1.14.2a3)
-
-  ## O que Mudou
-
-  ### Recursos
-  - Adicionar CLI de validação de deploy
-  - Melhorar a ergonomia de inicialização do LLM
-
-  ### Correções de Bugs
-  - Substituir pypdf e uv por versões corrigidas para CVE-2026-40260 e GHSA-pjjw-68hj-v9mw
-  - Atualizar requests para >=2.33.0 devido à vulnerabilidade de arquivo temporário CVE
-  - Preservar os argumentos de chamada da ferramenta Bedrock removendo o padrão truthy
-  - Sanitizar esquemas de ferramentas para modo estrito
-  - Remover flakiness do teste de serialização de embedding MemoryRecord
-
-  ### Documentação
-  - Limpar a linguagem do A2A empresarial
-  - Adicionar documentação de recursos do A2A empresarial
-  - Atualizar documentação do A2A OSS
-  - Atualizar changelog e versão para v1.14.2a2
-
-  ## Contribuidores
-
-  @Yanhu007, @greysonlalonde
-
-</Update>
-
 <Update label="10 abr 2026">
  ## v1.14.2a2

--- a/lib/crewai-files/src/crewai_files/init.py
+++ b/lib/crewai-files/src/crewai_files/init.py
@@ -152,4 +152,4 @@ __all__ = [
    "wrap_file_source",
 ]

-__version__ = "1.14.2rc1"
+__version__ = "1.14.2a2"
--- a/lib/crewai-tools/pyproject.toml
+++ b/lib/crewai-tools/pyproject.toml
@@ -10,7 +10,7 @@ requires-python = ">=3.10, <3.14"
 dependencies = [
    "pytube~=15.0.0",
    "requests>=2.33.0,<3",
-    "crewai==1.14.2rc1",
+    "crewai==1.14.2a2",
    "tiktoken~=0.8.0",
    "beautifulsoup4~=4.13.4",
    "python-docx~=1.2.0",
--- a/lib/crewai-tools/src/crewai_tools/init.py
+++ b/lib/crewai-tools/src/crewai_tools/init.py
@@ -305,4 +305,4 @@ __all__ = [
    "ZapierActionTools",
 ]

-__version__ = "1.14.2rc1"
+__version__ = "1.14.2a2"
--- a/lib/crewai/pyproject.toml
+++ b/lib/crewai/pyproject.toml
@@ -10,7 +10,7 @@ requires-python = ">=3.10, <3.14"
 dependencies = [
    # Core Dependencies
    "pydantic~=2.11.9",
-    "openai>=2.0.0,<3",
+    "openai>=1.83.0,<3",
    "instructor>=1.3.3",
    # Text Processing
    "pdfplumber~=0.11.4",
@@ -55,7 +55,7 @@ Repository = "https://github.com/crewAIInc/crewAI"

 [project.optional-dependencies]
 tools = [
-    "crewai-tools==1.14.2rc1",
+    "crewai-tools==1.14.2a2",
 ]
 embeddings = [
    "tiktoken~=0.8.0"
--- a/lib/crewai/src/crewai/init.py
+++ b/lib/crewai/src/crewai/init.py
@@ -46,7 +46,7 @@ def _suppress_pydantic_deprecation_warnings() -> None:

 _suppress_pydantic_deprecation_warnings()

-__version__ = "1.14.2rc1"
+__version__ = "1.14.2a2"
 _telemetry_submitted = False


--- a/lib/crewai/src/crewai/a2a/errors.py
+++ b/lib/crewai/src/crewai/a2a/errors.py
@@ -98,6 +98,7 @@ class A2AErrorCode(IntEnum):
    """The specified artifact was not found."""


+# Error code to default message mapping
 ERROR_MESSAGES: dict[int, str] = {
    A2AErrorCode.JSON_PARSE_ERROR: "Parse error",
    A2AErrorCode.INVALID_REQUEST: "Invalid Request",
--- a/lib/crewai/src/crewai/a2a/extensions/base.py
+++ b/lib/crewai/src/crewai/a2a/extensions/base.py
@@ -63,21 +63,25 @@ class A2AExtension(Protocol):
    Example:
        class MyExtension:
            def inject_tools(self, agent: Agent) -> None:
+                # Add custom tools to the agent
                pass

            def extract_state_from_history(
                self, conversation_history: Sequence[Message]
            ) -> ConversationState | None:
+                # Extract state from conversation
                return None

            def augment_prompt(
                self, base_prompt: str, conversation_state: ConversationState | None
            ) -> str:
+                # Add custom instructions
                return base_prompt

            def process_response(
                self, agent_response: Any, conversation_state: ConversationState | None
            ) -> Any:
+                # Modify response if needed
                return agent_response
    """

--- a/lib/crewai/src/crewai/a2a/utils/delegation.py
+++ b/lib/crewai/src/crewai/a2a/utils/delegation.py
@@ -47,6 +47,7 @@ from crewai.a2a.types import (
 )
 from crewai.a2a.updates import (
    PollingConfig,
+    PollingHandler,
    PushNotificationConfig,
    StreamingHandler,
    UpdateConfig,
@@ -586,6 +587,19 @@ async def _aexecute_a2a_delegation_impl(
    handler = get_handler(updates)
    use_polling = isinstance(updates, PollingConfig)

+    # If the user hasn't explicitly configured an updates strategy and the remote
+    # agent advertises that it does not support streaming, fall back to polling.
+    if updates is None and (
+        agent_card.capabilities is not None
+        and agent_card.capabilities.streaming is False
+    ):
+        logger.debug(
+            "Remote agent does not support streaming; falling back to PollingHandler",
+            extra={"endpoint": endpoint, "a2a_agent_name": a2a_agent_name},
+        )
+        handler = PollingHandler
+        use_polling = True
+
    handler_kwargs: dict[str, Any] = {
        "turn_number": turn_number,
        "is_multiturn": is_multiturn,
@@ -608,6 +622,18 @@ async def _aexecute_a2a_delegation_impl(
                "max_polls": updates.max_polls,
            }
        )
+    elif use_polling and updates is None:
+        # Fallback to polling because the agent card does not support streaming;
+        # use PollingConfig defaults so PollingHandler gets the kwargs it needs.
+        _default_polling = PollingConfig()
+        handler_kwargs.update(
+            {
+                "polling_interval": _default_polling.interval,
+                "polling_timeout": float(timeout),
+                "history_length": _default_polling.history_length,
+                "max_polls": _default_polling.max_polls,
+            }
+        )
    elif isinstance(updates, PushNotificationConfig):
        handler_kwargs.update(
            {
--- a/lib/crewai/src/crewai/a2a/utils/response_model.py
+++ b/lib/crewai/src/crewai/a2a/utils/response_model.py
@@ -77,6 +77,7 @@ def extract_a2a_agent_ids_from_config(
    else:
        configs = a2a_config

+    # Filter to only client configs (those with endpoint)
    client_configs: list[A2AClientConfigTypes] = [
        config for config in configs if isinstance(config, (A2AConfig, A2AClientConfig))
    ]
--- a/lib/crewai/src/crewai/agent/core.py
+++ b/lib/crewai/src/crewai/agent/core.py
@@ -1341,6 +1341,7 @@ class Agent(BaseAgent):

        raw_tools: list[BaseTool] = self.tools or []

+        # Inject memory tools for standalone kickoff (crew path handles its own)
        agent_memory = getattr(self, "memory", None)
        if agent_memory is not None:
            from crewai.tools.memory_tools import create_memory_tools
@@ -1398,6 +1399,7 @@ class Agent(BaseAgent):
        if input_files:
            all_files.update(input_files)

+        # Inject memory context for standalone kickoff (recall before execution)
        if agent_memory is not None:
            try:
                crewai_event_bus.emit(
@@ -1483,6 +1485,8 @@ class Agent(BaseAgent):
        Note:
            For explicit async usage outside of Flow, use kickoff_async() directly.
        """
+        # Magic auto-async: if inside event loop (e.g., inside a Flow),
+        # return coroutine for Flow to await
        if is_inside_event_loop():
            return self.kickoff_async(messages, response_format, input_files)

@@ -1633,7 +1637,7 @@ class Agent(BaseAgent):
                if isinstance(conversion_result, BaseModel):
                    formatted_result = conversion_result
            except ConverterError:
-                pass
+                pass  # Keep raw output if conversion fails
        else:
            raw_output = str(output) if not isinstance(output, str) else output

@@ -1715,6 +1719,7 @@ class Agent(BaseAgent):
        elif callable(self.guardrail):
            guardrail_callable = self.guardrail
        else:
+            # Should not happen if called from kickoff with guardrail check
            return output

        guardrail_result = process_guardrail(
--- a/lib/crewai/src/crewai/agent/planning_config.py
+++ b/lib/crewai/src/crewai/agent/planning_config.py
@@ -41,6 +41,7 @@ class PlanningConfig(BaseModel):
        from crewai import Agent
        from crewai.agent.planning_config import PlanningConfig

+        # Simple usage — fast, linear execution (default)
        agent = Agent(
            role="Researcher",
            goal="Research topics",
@@ -48,6 +49,7 @@ class PlanningConfig(BaseModel):
            planning_config=PlanningConfig(),
        )

+        # Balanced — replan only when steps fail
        agent = Agent(
            role="Researcher",
            goal="Research topics",
@@ -57,6 +59,7 @@ class PlanningConfig(BaseModel):
            ),
        )

+        # Full adaptive planning with refinement and replanning
        agent = Agent(
            role="Researcher",
            goal="Research topics",
@@ -66,7 +69,7 @@ class PlanningConfig(BaseModel):
                max_attempts=3,
                max_steps=10,
                plan_prompt="Create a focused plan for: {description}",
-                llm="gpt-4o-mini",
+                llm="gpt-4o-mini",  # Use cheaper model for planning
            ),
        )
        ```
--- a/lib/crewai/src/crewai/agent/utils.py
+++ b/lib/crewai/src/crewai/agent/utils.py
@@ -39,6 +39,7 @@ def handle_reasoning(agent: Agent, task: Task) -> None:
        agent: The agent performing the task.
        task: The task to execute.
    """
+    # Check if planning is enabled using the planning_enabled property
    if not getattr(agent, "planning_enabled", False):
        return

--- a/lib/crewai/src/crewai/agents/agent_adapters/openai_agents/openai_agent_tool_adapter.py
+++ b/lib/crewai/src/crewai/agents/agent_adapters/openai_agents/openai_agent_tool_adapter.py
@@ -99,10 +99,12 @@ class OpenAIAgentToolAdapter(BaseToolAdapter):
                Returns:
                    Tool execution result.
                """
+                # Get the parameter name from the schema
                param_name: str = next(
                    iter(tool.args_schema.model_json_schema()["properties"].keys())
                )

+                # Handle different argument types
                args_dict: dict[str, Any]
                if isinstance(arguments, dict):
                    args_dict = arguments
@@ -114,13 +116,16 @@ class OpenAIAgentToolAdapter(BaseToolAdapter):
                else:
                    args_dict = {param_name: str(arguments)}

+                # Run the tool with the processed arguments
                output: Any | Awaitable[Any] = tool._run(**args_dict)

+                # Await if the tool returned a coroutine
                if inspect.isawaitable(output):
                    result: Any = await output
                else:
                    result = output

+                # Ensure the result is JSON serializable
                if isinstance(result, (dict, list, str, int, float, bool, type(None))):
                    return result
                return str(result)
--- a/lib/crewai/src/crewai/agents/agent_builder/base_agent.py
+++ b/lib/crewai/src/crewai/agents/agent_builder/base_agent.py
@@ -383,6 +383,7 @@ class BaseAgent(BaseModel, ABC, metaclass=AgentMeta):
            if isinstance(tool, BaseTool):
                processed_tools.append(tool)
            elif all(hasattr(tool, attr) for attr in required_attrs):
+                # Tool has the required attributes, create a Tool instance
                processed_tools.append(Tool.from_langchain(tool))
            else:
                raise ValueError(
@@ -447,12 +448,14 @@ class BaseAgent(BaseModel, ABC, metaclass=AgentMeta):

    @model_validator(mode="after")
    def validate_and_set_attributes(self) -> Self:
+        # Validate required fields
        for field in ["role", "goal", "backstory"]:
            if getattr(self, field) is None:
                raise ValueError(
                    f"{field} must be provided either directly or through config"
                )

+        # Set private attributes
        self._logger = Logger(verbose=self.verbose)
        if self.max_rpm and not self._rpm_controller:
            self._rpm_controller = RPMController(
@@ -461,6 +464,7 @@ class BaseAgent(BaseModel, ABC, metaclass=AgentMeta):
        if not self._token_process:
            self._token_process = TokenProcess()

+        # Initialize security_config if not provided
        if self.security_config is None:
            self.security_config = SecurityConfig()

@@ -562,11 +566,14 @@ class BaseAgent(BaseModel, ABC, metaclass=AgentMeta):
            "actions",
        }

+        # Copy llm
        existing_llm = shallow_copy(self.llm)
        copied_knowledge = shallow_copy(self.knowledge)
        copied_knowledge_storage = shallow_copy(self.knowledge_storage)
+        # Properly copy knowledge sources if they exist
        existing_knowledge_sources = None
        if self.knowledge_sources:
+            # Create a shared storage instance for all knowledge sources
            shared_storage = (
                self.knowledge_sources[0].storage if self.knowledge_sources else None
            )
@@ -578,6 +585,7 @@ class BaseAgent(BaseModel, ABC, metaclass=AgentMeta):
                    if hasattr(source, "model_copy")
                    else shallow_copy(source)
                )
+                # Ensure all copied sources use the same storage instance
                copied_source.storage = shared_storage
                existing_knowledge_sources.append(copied_source)

--- a/lib/crewai/src/crewai/agents/constants.py
+++ b/lib/crewai/src/crewai/agents/constants.py
@@ -4,6 +4,8 @@ import re
 from typing import Final


+# crewai.agents.parser constants
+
 FINAL_ANSWER_ACTION: Final[str] = "Final Answer:"
 MISSING_ACTION_AFTER_THOUGHT_ERROR_MESSAGE: Final[str] = (
    "I did it wrong. Invalid Format: I missed the 'Action:' after 'Thought:'. I will do right next, and don't use a tool I have already used.\n"
--- a/lib/crewai/src/crewai/agents/crew_agent_executor.py
+++ b/lib/crewai/src/crewai/agents/crew_agent_executor.py
@@ -296,6 +296,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
        Returns:
            Final answer from the agent.
        """
+        # Check if model supports native function calling
        use_native_tools = (
            hasattr(self.llm, "supports_function_calling")
            and callable(getattr(self.llm, "supports_function_calling", None))
@@ -306,6 +307,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
        if use_native_tools:
            return self._invoke_loop_native_tools()

+        # Fall back to ReAct text-based pattern
        return self._invoke_loop_react()

    def _invoke_loop_react(self) -> AgentFinish:
@@ -345,6 +347,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
                    executor_context=self,
                    verbose=self.agent.verbose,
                )
+                # breakpoint()
                if self.response_model is not None:
                    try:
                        if isinstance(answer, BaseModel):
@@ -362,6 +365,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
                                text=answer,
                            )
                    except ValidationError:
+                        # If validation fails, convert BaseModel to JSON string for parsing
                        answer_str = (
                            answer.model_dump_json()
                            if isinstance(answer, BaseModel)
@@ -371,12 +375,14 @@ class CrewAgentExecutor(BaseAgentExecutor):
                            answer_str, self.use_stop_words
                        )  # type: ignore[assignment]
                else:
+                    # When no response_model, answer should be a string
                    answer_str = str(answer) if not isinstance(answer, str) else answer
                    formatted_answer = process_llm_response(
                        answer_str, self.use_stop_words
                    )  # type: ignore[assignment]

                if isinstance(formatted_answer, AgentAction):
+                    # Extract agent fingerprint if available
                    fingerprint_context = {}
                    if (
                        self.agent
@@ -420,6 +426,7 @@ class CrewAgentExecutor(BaseAgentExecutor):

            except Exception as e:
                if e.__class__.__module__.startswith("litellm"):
+                    # Do not retry on litellm errors
                    raise e
                if is_context_length_exceeded(e):
                    handle_context_length(
@@ -436,6 +443,10 @@ class CrewAgentExecutor(BaseAgentExecutor):
            finally:
                self.iterations += 1

+        # During the invoke loop, formatted_answer alternates between AgentAction
+        # (when the agent is using tools) and eventually becomes AgentFinish
+        # (when the agent reaches a final answer). This check confirms we've
+        # reached a final answer and helps type checking understand this transition.
        if not isinstance(formatted_answer, AgentFinish):
            raise RuntimeError(
                "Agent execution ended without reaching a final answer. "
@@ -454,7 +465,9 @@ class CrewAgentExecutor(BaseAgentExecutor):
        Returns:
            Final answer from the agent.
        """
+        # Convert tools to OpenAI schema format
        if not self.original_tools:
+            # No tools available, fall back to simple LLM call
            return self._invoke_loop_native_no_tools()

        openai_tools, available_functions, self._tool_name_mapping = (
@@ -477,6 +490,10 @@ class CrewAgentExecutor(BaseAgentExecutor):

                enforce_rpm_limit(self.request_within_rpm_limit)

+                # Call LLM with native tools
+                # Pass available_functions=None so the LLM returns tool_calls
+                # without executing them. The executor handles tool execution
+                # via _handle_native_tool_calls to properly manage message history.
                answer = get_llm_response(
                    llm=cast("BaseLLM", self.llm),
                    messages=self.messages,
@@ -491,26 +508,32 @@ class CrewAgentExecutor(BaseAgentExecutor):
                    verbose=self.agent.verbose,
                )

+                # Check if the response is a list of tool calls
                if (
                    isinstance(answer, list)
                    and answer
                    and self._is_tool_call_list(answer)
                ):
+                    # Handle tool calls - execute tools and add results to messages
                    tool_finish = self._handle_native_tool_calls(
                        answer, available_functions
                    )
+                    # If tool has result_as_answer=True, return immediately
                    if tool_finish is not None:
                        return tool_finish
+                    # Continue loop to let LLM analyze results and decide next steps
                    continue

+                # Text or other response - handle as potential final answer
                if isinstance(answer, str):
+                    # Text response - this is the final answer
                    formatted_answer = AgentFinish(
                        thought="",
                        output=answer,
                        text=answer,
                    )
                    self._invoke_step_callback(formatted_answer)
-                    self._append_message(answer)
+                    self._append_message(answer)  # Save final answer to messages
                    self._show_logs(formatted_answer)
                    return formatted_answer

@@ -526,13 +549,14 @@ class CrewAgentExecutor(BaseAgentExecutor):
                    self._show_logs(formatted_answer)
                    return formatted_answer

+                # Unexpected response type, treat as final answer
                formatted_answer = AgentFinish(
                    thought="",
                    output=str(answer),
                    text=str(answer),
                )
                self._invoke_step_callback(formatted_answer)
-                self._append_message(str(answer))
+                self._append_message(str(answer))  # Save final answer to messages
                self._show_logs(formatted_answer)
                return formatted_answer

@@ -603,10 +627,12 @@ class CrewAgentExecutor(BaseAgentExecutor):
        if not response:
            return False
        first_item = response[0]
+        # OpenAI-style
        if hasattr(first_item, "function") or (
            isinstance(first_item, dict) and "function" in first_item
        ):
            return True
+        # Anthropic-style (object with attributes)
        if (
            hasattr(first_item, "type")
            and getattr(first_item, "type", None) == "tool_use"
@@ -614,12 +640,14 @@ class CrewAgentExecutor(BaseAgentExecutor):
            return True
        if hasattr(first_item, "name") and hasattr(first_item, "input"):
            return True
+        # Bedrock-style (dict with name and input keys)
        if (
            isinstance(first_item, dict)
            and "name" in first_item
            and "input" in first_item
        ):
            return True
+        # Gemini-style
        if hasattr(first_item, "function_call") and first_item.function_call:
            return True
        return False
@@ -678,6 +706,8 @@ class CrewAgentExecutor(BaseAgentExecutor):
                for _, func_name, _ in parsed_calls
            )

+            # Preserve historical sequential behavior for result_as_answer batches.
+            # Also avoid threading around usage counters for max_usage_count tools.
            if has_result_as_answer_in_batch or has_max_usage_count_in_batch:
                logger.debug(
                    "Skipping parallel native execution because batch includes result_as_answer or max_usage_count tool"
@@ -743,6 +773,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
                self.messages.append(reasoning_message)
                return None

+        # Sequential behavior: process only first tool call, then force reflection.
        call_id, func_name, func_args = parsed_calls[0]
        self._append_assistant_tool_calls_message([(call_id, func_name, func_args)])

@@ -1171,6 +1202,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
                                text=answer,
                            )
                    except ValidationError:
+                        # If validation fails, convert BaseModel to JSON string for parsing
                        answer_str = (
                            answer.model_dump_json()
                            if isinstance(answer, BaseModel)
@@ -1180,6 +1212,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
                            answer_str, self.use_stop_words
                        )  # type: ignore[assignment]
                else:
+                    # When no response_model, answer should be a string
                    answer_str = str(answer) if not isinstance(answer, str) else answer
                    formatted_answer = process_llm_response(
                        answer_str, self.use_stop_words
@@ -1286,6 +1319,10 @@ class CrewAgentExecutor(BaseAgentExecutor):

                enforce_rpm_limit(self.request_within_rpm_limit)

+                # Call LLM with native tools
+                # Pass available_functions=None so the LLM returns tool_calls
+                # without executing them. The executor handles tool execution
+                # via _handle_native_tool_calls to properly manage message history.
                answer = await aget_llm_response(
                    llm=cast("BaseLLM", self.llm),
                    messages=self.messages,
@@ -1299,26 +1336,32 @@ class CrewAgentExecutor(BaseAgentExecutor):
                    executor_context=self,
                    verbose=self.agent.verbose,
                )
+                # Check if the response is a list of tool calls
                if (
                    isinstance(answer, list)
                    and answer
                    and self._is_tool_call_list(answer)
                ):
+                    # Handle tool calls - execute tools and add results to messages
                    tool_finish = self._handle_native_tool_calls(
                        answer, available_functions
                    )
+                    # If tool has result_as_answer=True, return immediately
                    if tool_finish is not None:
                        return tool_finish
+                    # Continue loop to let LLM analyze results and decide next steps
                    continue

+                # Text or other response - handle as potential final answer
                if isinstance(answer, str):
+                    # Text response - this is the final answer
                    formatted_answer = AgentFinish(
                        thought="",
                        output=answer,
                        text=answer,
                    )
                    await self._ainvoke_step_callback(formatted_answer)
-                    self._append_message(answer)
+                    self._append_message(answer)  # Save final answer to messages
                    self._show_logs(formatted_answer)
                    return formatted_answer

@@ -1334,13 +1377,14 @@ class CrewAgentExecutor(BaseAgentExecutor):
                    self._show_logs(formatted_answer)
                    return formatted_answer

+                # Unexpected response type, treat as final answer
                formatted_answer = AgentFinish(
                    thought="",
                    output=str(answer),
                    text=str(answer),
                )
                await self._ainvoke_step_callback(formatted_answer)
-                self._append_message(str(answer))
+                self._append_message(str(answer))  # Save final answer to messages
                self._show_logs(formatted_answer)
                return formatted_answer

@@ -1411,6 +1455,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
        Returns:
            Updated action or final answer.
        """
+        # Special case for add_image_tool
        add_image_tool = I18N_DEFAULT.tools("add_image")
        if (
            isinstance(add_image_tool, dict)
@@ -1530,14 +1575,17 @@ class CrewAgentExecutor(BaseAgentExecutor):
        training_handler = CrewTrainingHandler(TRAINING_DATA_FILE)
        training_data = training_handler.load() or {}

+        # Initialize or retrieve agent's training data
        agent_training_data = training_data.get(agent_id, {})

        if human_feedback is not None:
+            # Save initial output and human feedback
            agent_training_data[train_iteration] = {
                "initial_output": result.output,
                "human_feedback": human_feedback,
            }
        else:
+            # Save improved output
            if train_iteration in agent_training_data:
                agent_training_data[train_iteration]["improved_output"] = result.output
            else:
@@ -1551,6 +1599,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
                    )
                return

+        # Update the training data and save
        training_data[agent_id] = agent_training_data
        training_handler.save(training_data)

--- a/lib/crewai/src/crewai/agents/parser.py
+++ b/lib/crewai/src/crewai/agents/parser.py
@@ -94,8 +94,11 @@ def parse(text: str) -> AgentAction | AgentFinish:

    if includes_answer:
        final_answer = text.split(FINAL_ANSWER_ACTION)[-1].strip()
+        # Check whether the final answer ends with triple backticks.
        if final_answer.endswith("```"):
+            # Count occurrences of triple backticks in the final answer.
            count = final_answer.count("```")
+            # If count is odd then it's an unmatched trailing set; remove it.
            if count % 2 != 0:
                final_answer = final_answer[:-3].rstrip()
        return AgentFinish(thought=thought, output=final_answer, text=text)
@@ -143,6 +146,7 @@ def _extract_thought(text: str) -> str:
    if thought_index == -1:
        return ""
    thought = text[:thought_index].strip()
+    # Remove any triple backticks from the thought string
    return thought.replace("```", "").strip()


@@ -167,9 +171,18 @@ def _safe_repair_json(tool_input: str) -> str:
    Returns:
        The repaired JSON string or original if repair fails.
    """
+    # Skip repair if the input starts and ends with square brackets
+    # Explanation: The JSON parser has issues handling inputs that are enclosed in square brackets ('[]').
+    # These are typically valid JSON arrays or strings that do not require repair. Attempting to repair such inputs
+    # might lead to unintended alterations, such as wrapping the entire input in additional layers or modifying
+    # the structure in a way that changes its meaning. By skipping the repair for inputs that start and end with
+    # square brackets, we preserve the integrity of these valid JSON structures and avoid unnecessary modifications.
    if tool_input.startswith("[") and tool_input.endswith("]"):
        return tool_input

+    # Before repair, handle common LLM issues:
+    # 1. Replace """ with " to avoid JSON parser errors
+
    tool_input = tool_input.replace('"""', '"')

    result = repair_json(tool_input)
--- a/lib/crewai/src/crewai/agents/planner_observer.py
+++ b/lib/crewai/src/crewai/agents/planner_observer.py
@@ -83,6 +83,10 @@ class PlannerObserver:
            return create_llm(config.llm)
        return self.agent.llm

+    # ------------------------------------------------------------------
+    # Public API
+    # ------------------------------------------------------------------
+
    def observe(
        self,
        completed_step: TodoItem,
@@ -178,6 +182,9 @@ class PlannerObserver:
                ),
            )

+            # Don't force a full replan — the step may have succeeded even if the
+            # observer LLM failed to parse the result. Defaulting to "continue" is
+            # far less disruptive than wiping the entire plan on every observer error.
            return StepObservation(
                step_completed_successfully=True,
                key_information_learned="",
@@ -214,6 +221,10 @@ class PlannerObserver:

        return remaining_todos

+    # ------------------------------------------------------------------
+    # Internal: Message building
+    # ------------------------------------------------------------------
+
    def _build_observation_messages(
        self,
        completed_step: TodoItem,
@@ -228,11 +239,15 @@ class PlannerObserver:
            task_desc = self.task.description or ""
            task_goal = self.task.expected_output or ""
        elif self.kickoff_input:
+            # Standalone kickoff path — no Task object, but we have the raw input.
+            # Extract just the ## Task section so the observer sees the actual goal,
+            # not the full enriched instruction with env/tools/verification noise.
            task_desc = extract_task_section(self.kickoff_input)
            task_goal = "Complete the task successfully"

        system_prompt = I18N_DEFAULT.retrieve("planning", "observation_system_prompt")

+        # Build context of what's been done
        completed_summary = ""
        if all_completed:
            completed_lines = []
@@ -246,6 +261,7 @@ class PlannerObserver:
                completed_lines
            )

+        # Build remaining plan
        remaining_summary = ""
        if remaining_todos:
            remaining_lines = [
@@ -290,14 +306,17 @@ class PlannerObserver:
        if isinstance(response, StepObservation):
            return response

+        # JSON string path — most common miss before this fix
        if isinstance(response, str):
            text = response.strip()
            try:
                return StepObservation.model_validate_json(text)
            except Exception:  # noqa: S110
                pass
+            # Some LLMs wrap the JSON in markdown fences
            if text.startswith("```"):
                lines = text.split("\n")
+                # Strip first and last lines (``` markers)
                inner = "\n".join(
                    lines[1:-1] if lines[-1].strip() == "```" else lines[1:]
                )
@@ -306,12 +325,14 @@ class PlannerObserver:
                except Exception:  # noqa: S110
                    pass

+        # Dict path
        if isinstance(response, dict):
            try:
                return StepObservation.model_validate(response)
            except Exception:  # noqa: S110
                pass

+        # Last resort — log what we got so it's diagnosable
        logger.warning(
            "Could not parse observation response (type=%s). "
            "Falling back to default failure observation. Preview: %.200s",
--- a/lib/crewai/src/crewai/agents/step_executor.py
+++ b/lib/crewai/src/crewai/agents/step_executor.py
@@ -108,6 +108,7 @@ class StepExecutor:
        self.request_within_rpm_limit = request_within_rpm_limit
        self.callbacks = callbacks or []

+        # Native tool support — set up once
        self._use_native_tools = check_native_tool_support(
            self.llm, self.original_tools
        )
@@ -120,6 +121,10 @@ class StepExecutor:
                _,
            ) = setup_native_tools(self.original_tools)

+    # ------------------------------------------------------------------
+    # Public API
+    # ------------------------------------------------------------------
+
    def execute(
        self,
        todo: TodoItem,
@@ -185,6 +190,10 @@ class StepExecutor:
                execution_time=elapsed,
            )

+    # ------------------------------------------------------------------
+    # Internal: Message building
+    # ------------------------------------------------------------------
+
    def _build_isolated_messages(
        self, todo: TodoItem, context: StepExecutionContext
    ) -> list[LLMMessage]:
@@ -228,6 +237,10 @@ class StepExecutor:
        """Build the user prompt for this specific step."""
        parts: list[str] = []

+        # Include overall task context so the executor knows the full goal and
+        # required output format/location — critical for knowing WHAT to produce.
+        # We extract only the task body (not tool instructions or verification
+        # sections) to avoid duplicating directives already in the system prompt.
        if context.task_description:
            task_section = extract_task_section(context.task_description)
            if task_section:
@@ -254,6 +267,7 @@ class StepExecutor:
                )
            )

+        # Include dependency results (final results only, no traces)
        if context.dependency_results:
            parts.append(
                I18N_DEFAULT.retrieve("planning", "step_executor_context_header")
@@ -269,6 +283,10 @@ class StepExecutor:

        return "\n".join(parts)

+    # ------------------------------------------------------------------
+    # Internal: Multi-turn execution loop
+    # ------------------------------------------------------------------
+
    def _execute_text_parsed(
        self,
        messages: list[LLMMessage],
@@ -288,6 +306,7 @@ class StepExecutor:
        last_tool_result = ""

        for _ in range(max_step_iterations):
+            # Check step timeout
            if step_timeout and start_time:
                elapsed = time.monotonic() - start_time
                if elapsed >= step_timeout:
@@ -312,12 +331,17 @@ class StepExecutor:
                tool_calls_made.append(formatted.tool)
                tool_result = self._execute_text_tool_with_events(formatted)
                last_tool_result = tool_result
+                # Append the assistant's reasoning + action, then the observation.
+                # _build_observation_message handles vision sentinels so the LLM
+                # receives an image content block instead of raw base64 text.
                messages.append({"role": "assistant", "content": answer_str})
                messages.append(self._build_observation_message(tool_result))
                continue

+            # Raw text response with no Final Answer marker — treat as done
            return answer_str

+        # Max iterations reached — return the last tool result we accumulated
        return last_tool_result

    def _execute_text_tool_with_events(self, formatted: AgentAction) -> str:
@@ -405,6 +429,10 @@ class StepExecutor:
                return {"input": stripped_input}
        return {"input": str(tool_input)}

+    # ------------------------------------------------------------------
+    # Internal: Vision support
+    # ------------------------------------------------------------------
+
    @staticmethod
    def _parse_vision_sentinel(raw: str) -> tuple[str, str] | None:
        """Parse a VISION_IMAGE sentinel into (media_type, base64_data), or None."""
@@ -489,6 +517,7 @@ class StepExecutor:
        accumulated_results: list[str] = []

        for _ in range(max_step_iterations):
+            # Check step timeout
            if step_timeout and start_time:
                elapsed = time.monotonic() - start_time
                if elapsed >= step_timeout:
@@ -512,14 +541,19 @@ class StepExecutor:
                return answer.model_dump_json()

            if isinstance(answer, list) and answer and is_tool_call_list(answer):
+                # _execute_native_tool_calls appends assistant + tool messages
+                # to `messages` as a side-effect, so the next LLM call will
+                # see the full conversation history including tool outputs.
                result = self._execute_native_tool_calls(
                    answer, messages, tool_calls_made
                )
                accumulated_results.append(result)
                continue

+            # Text answer → LLM decided the step is done
            return str(answer)

+        # Max iterations reached — return everything we accumulated
        return "\n".join(filter(None, accumulated_results))

    def _execute_native_tool_calls(
@@ -565,6 +599,9 @@ class StepExecutor:
                    parsed = self._parse_vision_sentinel(raw_content)
                    if parsed:
                        media_type, b64_data = parsed
+                        # Replace the sentinel with a standard image_url content block.
+                        # Each provider's _format_messages handles conversion to
+                        # its native format (e.g. Anthropic image blocks).
                        modified: LLMMessage = cast(
                            LLMMessage, dict(call_result.tool_message)
                        )
--- a/lib/crewai/src/crewai/cli/templates/crew/pyproject.toml
+++ b/lib/crewai/src/crewai/cli/templates/crew/pyproject.toml
@@ -5,7 +5,7 @@ description = "{{name}} using crewAI"
 authors = [{ name = "Your Name", email = "you@example.com" }]
 requires-python = ">=3.10,<3.14"
 dependencies = [
-    "crewai[tools]==1.14.2rc1"
+    "crewai[tools]==1.14.2a2"
 ]

 [project.scripts]
--- a/lib/crewai/src/crewai/cli/templates/flow/pyproject.toml
+++ b/lib/crewai/src/crewai/cli/templates/flow/pyproject.toml
@@ -5,7 +5,7 @@ description = "{{name}} using crewAI"
 authors = [{ name = "Your Name", email = "you@example.com" }]
 requires-python = ">=3.10,<3.14"
 dependencies = [
-    "crewai[tools]==1.14.2rc1"
+    "crewai[tools]==1.14.2a2"
 ]

 [project.scripts]
--- a/lib/crewai/src/crewai/cli/templates/tool/pyproject.toml
+++ b/lib/crewai/src/crewai/cli/templates/tool/pyproject.toml
@@ -5,7 +5,7 @@ description = "Power up your crews with {{folder_name}}"
 readme = "README.md"
 requires-python = ">=3.10,<3.14"
 dependencies = [
-    "crewai[tools]==1.14.2rc1"
+    "crewai[tools]==1.14.2a2"
 ]

 [tool.crewai]
--- a/lib/crewai/src/crewai/lite_agent.py
+++ b/lib/crewai/src/crewai/lite_agent.py
@@ -16,6 +16,7 @@ from typing import (
    get_origin,
 )
 import uuid
+import warnings

 from pydantic import (
    UUID4,
@@ -25,7 +26,7 @@ from pydantic import (
    field_validator,
    model_validator,
 )
-from typing_extensions import Self, deprecated
+from typing_extensions import Self


 if TYPE_CHECKING:
@@ -172,12 +173,9 @@ def _kickoff_with_a2a_support(
    )


-@deprecated(
-    "LiteAgent is deprecated and will be removed in v2.0.0.",
-    category=FutureWarning,
-)
 class LiteAgent(FlowTrackable, BaseModel):
-    """A lightweight agent that can process messages and use tools.
+    """
+    A lightweight agent that can process messages and use tools.

    .. deprecated::
        LiteAgent is deprecated and will be removed in a future version.
@@ -280,6 +278,18 @@ class LiteAgent(FlowTrackable, BaseModel):
    )
    _memory: Any = PrivateAttr(default=None)

+    @model_validator(mode="after")
+    def emit_deprecation_warning(self) -> Self:
+        """Emit deprecation warning for LiteAgent usage."""
+        warnings.warn(
+            "LiteAgent is deprecated and will be removed in a future version. "
+            "Use Agent().kickoff(messages) instead, which provides the same "
+            "functionality with additional features like memory and knowledge support.",
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        return self
+
    @model_validator(mode="after")
    def setup_llm(self) -> Self:
        """Set up the LLM and other components after initialization."""
--- a/lib/crewai/src/crewai/llm.py
+++ b/lib/crewai/src/crewai/llm.py
@@ -51,7 +51,6 @@ from crewai.utilities.exceptions.context_window_exceeding_exception import (
 )
 from crewai.utilities.logger_utils import suppress_warnings
 from crewai.utilities.string_utils import sanitize_tool_name
-from crewai.utilities.token_counter_callback import TokenCalcHandler


 try:
@@ -76,13 +75,8 @@ try:
    from litellm.types.utils import (
        ChatCompletionDeltaToolCall,
        Choices,
-        Delta as LiteLLMDelta,
        Function,
-        Message,
        ModelResponse,
-        ModelResponseBase,
-        ModelResponseStream,
-        StreamingChoices as LiteLLMStreamingChoices,
    )
    from litellm.utils import supports_response_schema

@@ -91,11 +85,6 @@ except ImportError:
    LITELLM_AVAILABLE = False
    litellm = None  # type: ignore[assignment]
    Choices = None  # type: ignore[assignment, misc]
-    LiteLLMDelta = None  # type: ignore[assignment, misc]
-    Message = None  # type: ignore[assignment, misc]
-    ModelResponseBase = None  # type: ignore[assignment, misc]
-    ModelResponseStream = None  # type: ignore[assignment, misc]
-    LiteLLMStreamingChoices = None  # type: ignore[assignment, misc]
    get_supported_openai_params = None  # type: ignore[assignment]
    ChatCompletionDeltaToolCall = None  # type: ignore[assignment, misc]
    Function = None  # type: ignore[assignment, misc]
@@ -720,7 +709,7 @@ class LLM(BaseLLM):
                chunk_content = None
                response_id = None

-                if isinstance(chunk, ModelResponseBase):
+                if hasattr(chunk, "id"):
                    response_id = chunk.id

                # Safely extract content from various chunk formats
@@ -729,16 +718,18 @@ class LLM(BaseLLM):
                    choices = None
                    if isinstance(chunk, dict) and "choices" in chunk:
                        choices = chunk["choices"]
-                    elif isinstance(chunk, ModelResponseStream):
-                        choices = chunk.choices
+                    elif hasattr(chunk, "choices"):
+                        # Check if choices is not a type but an actual attribute with value
+                        if not isinstance(chunk.choices, type):
+                            choices = chunk.choices

                    # Try to extract usage information if available
-                    # NOTE: usage is a pydantic extra field on ModelResponseBase,
-                    # so it must be accessed via model_extra.
                    if isinstance(chunk, dict) and "usage" in chunk:
                        usage_info = chunk["usage"]
-                    elif isinstance(chunk, ModelResponseBase) and chunk.model_extra:
-                        usage_info = chunk.model_extra.get("usage") or usage_info
+                    elif hasattr(chunk, "usage"):
+                        # Check if usage is not a type but an actual attribute with value
+                        if not isinstance(chunk.usage, type):
+                            usage_info = chunk.usage

                    if choices and len(choices) > 0:
                        choice = choices[0]
@@ -747,7 +738,7 @@ class LLM(BaseLLM):
                        delta = None
                        if isinstance(choice, dict) and "delta" in choice:
                            delta = choice["delta"]
-                        elif isinstance(choice, LiteLLMStreamingChoices):
+                        elif hasattr(choice, "delta"):
                            delta = choice.delta

                        # Extract content from delta
@@ -757,7 +748,7 @@ class LLM(BaseLLM):
                                if "content" in delta and delta["content"] is not None:
                                    chunk_content = delta["content"]
                            # Handle object format
-                            elif isinstance(delta, LiteLLMDelta):
+                            elif hasattr(delta, "content"):
                                chunk_content = delta.content

                            # Handle case where content might be None or empty
@@ -830,8 +821,9 @@ class LLM(BaseLLM):
                        choices = None
                        if isinstance(last_chunk, dict) and "choices" in last_chunk:
                            choices = last_chunk["choices"]
-                        elif isinstance(last_chunk, ModelResponseStream):
-                            choices = last_chunk.choices
+                        elif hasattr(last_chunk, "choices"):
+                            if not isinstance(last_chunk.choices, type):
+                                choices = last_chunk.choices

                        if choices and len(choices) > 0:
                            choice = choices[0]
@@ -840,14 +832,14 @@ class LLM(BaseLLM):
                            message = None
                            if isinstance(choice, dict) and "message" in choice:
                                message = choice["message"]
-                            elif isinstance(choice, Choices):
+                            elif hasattr(choice, "message"):
                                message = choice.message

                            if message:
                                content = None
                                if isinstance(message, dict) and "content" in message:
                                    content = message["content"]
-                                elif isinstance(message, Message):
+                                elif hasattr(message, "content"):
                                    content = message.content

                                if content:
@@ -874,23 +866,24 @@ class LLM(BaseLLM):
                    choices = None
                    if isinstance(last_chunk, dict) and "choices" in last_chunk:
                        choices = last_chunk["choices"]
-                    elif isinstance(last_chunk, ModelResponseStream):
-                        choices = last_chunk.choices
+                    elif hasattr(last_chunk, "choices"):
+                        if not isinstance(last_chunk.choices, type):
+                            choices = last_chunk.choices

                    if choices and len(choices) > 0:
                        choice = choices[0]

-                        delta = None
-                        if isinstance(choice, dict) and "delta" in choice:
-                            delta = choice["delta"]
-                        elif isinstance(choice, LiteLLMStreamingChoices):
-                            delta = choice.delta
+                        message = None
+                        if isinstance(choice, dict) and "message" in choice:
+                            message = choice["message"]
+                        elif hasattr(choice, "message"):
+                            message = choice.message

-                        if delta:
-                            if isinstance(delta, dict) and "tool_calls" in delta:
-                                tool_calls = delta["tool_calls"]
-                            elif isinstance(delta, LiteLLMDelta):
-                                tool_calls = delta.tool_calls
+                        if message:
+                            if isinstance(message, dict) and "tool_calls" in message:
+                                tool_calls = message["tool_calls"]
+                            elif hasattr(message, "tool_calls"):
+                                tool_calls = message.tool_calls
            except Exception as e:
                logging.debug(f"Error checking for tool calls: {e}")

@@ -1044,7 +1037,7 @@ class LLM(BaseLLM):
        """
        if callbacks and len(callbacks) > 0:
            for callback in callbacks:
-                if isinstance(callback, TokenCalcHandler):
+                if hasattr(callback, "log_success_event"):
                    # Use the usage_info we've been tracking
                    if not usage_info:
                        # Try to get usage from the last chunk if we haven't already
@@ -1055,14 +1048,9 @@ class LLM(BaseLLM):
                                    and "usage" in last_chunk
                                ):
                                    usage_info = last_chunk["usage"]
-                                elif (
-                                    isinstance(last_chunk, ModelResponseBase)
-                                    and last_chunk.model_extra
-                                ):
-                                    usage_info = (
-                                        last_chunk.model_extra.get("usage")
-                                        or usage_info
-                                    )
+                                elif hasattr(last_chunk, "usage"):
+                                    if not isinstance(last_chunk.usage, type):
+                                        usage_info = last_chunk.usage
                        except Exception as e:
                            logging.debug(f"Error extracting usage info: {e}")

@@ -1135,10 +1123,13 @@ class LLM(BaseLLM):
                params["response_model"] = response_model
            response = litellm.completion(**params)

-            if isinstance(response, ModelResponseBase) and response.model_extra:
-                usage_info = response.model_extra.get("usage")
-                if usage_info:
-                    self._track_token_usage_internal(usage_info)
+            if (
+                hasattr(response, "usage")
+                and not isinstance(response.usage, type)
+                and response.usage
+            ):
+                usage_info = response.usage
+                self._track_token_usage_internal(usage_info)

        except LLMContextLengthExceededError:
            # Re-raise our own context length error
@@ -1150,11 +1141,7 @@ class LLM(BaseLLM):
                raise LLMContextLengthExceededError(error_msg) from e
            raise

-        response_usage = self._usage_to_dict(
-            response.model_extra.get("usage")
-            if isinstance(response, ModelResponseBase) and response.model_extra
-            else None
-        )
+        response_usage = self._usage_to_dict(getattr(response, "usage", None))

        # --- 2) Handle structured output response (when response_model is provided)
        if response_model is not None:
@@ -1179,13 +1166,8 @@ class LLM(BaseLLM):
        # --- 3) Handle callbacks with usage info
        if callbacks and len(callbacks) > 0:
            for callback in callbacks:
-                if isinstance(callback, TokenCalcHandler):
-                    usage_info = (
-                        response.model_extra.get("usage")
-                        if isinstance(response, ModelResponseBase)
-                        and response.model_extra
-                        else None
-                    )
+                if hasattr(callback, "log_success_event"):
+                    usage_info = getattr(response, "usage", None)
                    if usage_info:
                        callback.log_success_event(
                            kwargs=params,
@@ -1194,7 +1176,7 @@ class LLM(BaseLLM):
                            end_time=0,
                        )
        # --- 4) Check for tool calls
-        tool_calls = response_message.tool_calls or []
+        tool_calls = getattr(response_message, "tool_calls", [])

        # --- 5) If no tool calls or no available functions, return the text response directly as long as there is a text response
        if (not tool_calls or not available_functions) and text_response:
@@ -1287,10 +1269,13 @@ class LLM(BaseLLM):
                params["response_model"] = response_model
            response = await litellm.acompletion(**params)

-            if isinstance(response, ModelResponseBase) and response.model_extra:
-                usage_info = response.model_extra.get("usage")
-                if usage_info:
-                    self._track_token_usage_internal(usage_info)
+            if (
+                hasattr(response, "usage")
+                and not isinstance(response.usage, type)
+                and response.usage
+            ):
+                usage_info = response.usage
+                self._track_token_usage_internal(usage_info)

        except LLMContextLengthExceededError:
            # Re-raise our own context length error
@@ -1302,11 +1287,7 @@ class LLM(BaseLLM):
                raise LLMContextLengthExceededError(error_msg) from e
            raise

-        response_usage = self._usage_to_dict(
-            response.model_extra.get("usage")
-            if isinstance(response, ModelResponseBase) and response.model_extra
-            else None
-        )
+        response_usage = self._usage_to_dict(getattr(response, "usage", None))

        if response_model is not None:
            if isinstance(response, BaseModel):
@@ -1328,13 +1309,8 @@ class LLM(BaseLLM):

        if callbacks and len(callbacks) > 0:
            for callback in callbacks:
-                if isinstance(callback, TokenCalcHandler):
-                    usage_info = (
-                        response.model_extra.get("usage")
-                        if isinstance(response, ModelResponseBase)
-                        and response.model_extra
-                        else None
-                    )
+                if hasattr(callback, "log_success_event"):
+                    usage_info = getattr(response, "usage", None)
                    if usage_info:
                        callback.log_success_event(
                            kwargs=params,
@@ -1343,7 +1319,7 @@ class LLM(BaseLLM):
                            end_time=0,
                        )

-        tool_calls = response_message.tool_calls or []
+        tool_calls = getattr(response_message, "tool_calls", [])

        if (not tool_calls or not available_functions) and text_response:
            self._handle_emit_call_events(
@@ -1418,19 +1394,18 @@ class LLM(BaseLLM):
            async for chunk in await litellm.acompletion(**params):
                chunk_count += 1
                chunk_content = None
-                response_id = chunk.id if isinstance(chunk, ModelResponseBase) else None
+                response_id = chunk.id if hasattr(chunk, "id") else None

                try:
                    choices = None
                    if isinstance(chunk, dict) and "choices" in chunk:
                        choices = chunk["choices"]
-                    elif isinstance(chunk, ModelResponseStream):
-                        choices = chunk.choices
+                    elif hasattr(chunk, "choices"):
+                        if not isinstance(chunk.choices, type):
+                            choices = chunk.choices

-                    if isinstance(chunk, ModelResponseBase) and chunk.model_extra:
-                        chunk_usage = chunk.model_extra.get("usage")
-                        if chunk_usage is not None:
-                            usage_info = chunk_usage
+                    if hasattr(chunk, "usage") and chunk.usage is not None:
+                        usage_info = chunk.usage

                    if choices and len(choices) > 0:
                        first_choice = choices[0]
@@ -1438,19 +1413,19 @@ class LLM(BaseLLM):

                        if isinstance(first_choice, dict):
                            delta = first_choice.get("delta", {})
-                        elif isinstance(first_choice, LiteLLMStreamingChoices):
+                        elif hasattr(first_choice, "delta"):
                            delta = first_choice.delta

                        if delta:
                            if isinstance(delta, dict):
                                chunk_content = delta.get("content")
-                            elif isinstance(delta, LiteLLMDelta):
+                            elif hasattr(delta, "content"):
                                chunk_content = delta.content

                            tool_calls: list[ChatCompletionDeltaToolCall] | None = None
                            if isinstance(delta, dict):
                                tool_calls = delta.get("tool_calls")
-                            elif isinstance(delta, LiteLLMDelta):
+                            elif hasattr(delta, "tool_calls"):
                                tool_calls = delta.tool_calls

                            if tool_calls:
@@ -1486,7 +1461,7 @@ class LLM(BaseLLM):

            if callbacks and len(callbacks) > 0 and usage_info:
                for callback in callbacks:
-                    if isinstance(callback, TokenCalcHandler):
+                    if hasattr(callback, "log_success_event"):
                        callback.log_success_event(
                            kwargs=params,
                            response_obj={"usage": usage_info},
@@ -1945,7 +1920,7 @@ class LLM(BaseLLM):
            return None
        if isinstance(usage, dict):
            return usage
-        if isinstance(usage, BaseModel):
+        if hasattr(usage, "model_dump"):
            result: dict[str, Any] = usage.model_dump()
            return result
        if hasattr(usage, "__dict__"):
@@ -2009,7 +1984,7 @@ class LLM(BaseLLM):
                )
            return messages

-        provider = self.provider or self.model
+        provider = getattr(self, "provider", None) or self.model

        for msg in messages:
            files = msg.get("files")
@@ -2060,7 +2035,7 @@ class LLM(BaseLLM):
                )
            return messages

-        provider = self.provider or self.model
+        provider = getattr(self, "provider", None) or self.model

        for msg in messages:
            files = msg.get("files")
--- a/lib/crewai/src/crewai/llms/providers/bedrock/completion.py
+++ b/lib/crewai/src/crewai/llms/providers/bedrock/completion.py
@@ -17,7 +17,10 @@ from crewai.utilities.agent_utils import is_context_length_exceeded
 from crewai.utilities.exceptions.context_window_exceeding_exception import (
    LLMContextLengthExceededError,
 )
-from crewai.utilities.pydantic_schema_utils import generate_model_description
+from crewai.utilities.pydantic_schema_utils import (
+    generate_model_description,
+    sanitize_tool_params_for_bedrock_strict,
+)
 from crewai.utilities.types import LLMMessage


@@ -170,6 +173,7 @@ class ToolSpec(TypedDict, total=False):
    name: Required[str]
    description: Required[str]
    inputSchema: ToolInputSchema
+    strict: bool


 class ConverseToolTypeDef(TypedDict):
@@ -1984,10 +1988,21 @@ class BedrockCompletion(BaseLLM):
                    "description": description,
                }

+                func_info = tool.get("function", {})
+                strict_enabled = bool(func_info.get("strict"))
+
                if parameters and isinstance(parameters, dict):
-                    input_schema: ToolInputSchema = {"json": parameters}
+                    schema_params = (
+                        sanitize_tool_params_for_bedrock_strict(parameters)
+                        if strict_enabled
+                        else parameters
+                    )
+                    input_schema: ToolInputSchema = {"json": schema_params}
                    tool_spec["inputSchema"] = input_schema

+                if strict_enabled:
+                    tool_spec["strict"] = True
+
                converse_tool: ConverseToolTypeDef = {"toolSpec": tool_spec}

                converse_tools.append(converse_tool)
--- a/lib/crewai/src/crewai/mcp/tool_resolver.py
+++ b/lib/crewai/src/crewai/mcp/tool_resolver.py
@@ -417,18 +417,9 @@ class MCPToolResolver:

                args_schema = None
                if tool_def.get("inputSchema"):
-                    try:
-                        args_schema = self._json_schema_to_pydantic(
-                            tool_name, tool_def["inputSchema"]
-                        )
-                    except Exception as e:
-                        self._logger.log(
-                            "warning",
-                            f"Failed to build args schema for MCP tool "
-                            f"'{tool_name}': {e}. Registering tool without a "
-                            "typed schema.",
-                        )
-                        args_schema = None
+                    args_schema = self._json_schema_to_pydantic(
+                        tool_name, tool_def["inputSchema"]
+                    )

                tool_schema = {
                    "description": tool_def.get("description", ""),
--- a/lib/crewai/src/crewai/task.py
+++ b/lib/crewai/src/crewai/task.py
@@ -45,7 +45,6 @@ from crewai.events.types.task_events import (
    TaskStartedEvent,
 )
 from crewai.llms.base_llm import BaseLLM
-from crewai.llms.providers.openai.completion import OpenAICompletion
 from crewai.security import Fingerprint, SecurityConfig
 from crewai.tasks.output_format import OutputFormat
 from crewai.tasks.task_output import TaskOutput
@@ -302,14 +301,12 @@ class Task(BaseModel):

    @model_validator(mode="after")
    def validate_required_fields(self) -> Self:
-        if self.description is None:
-            raise ValueError(
-                "description must be provided either directly or through config"
-            )
-        if self.expected_output is None:
-            raise ValueError(
-                "expected_output must be provided either directly or through config"
-            )
+        required_fields = ["description", "expected_output"]
+        for field in required_fields:
+            if getattr(self, field) is None:
+                raise ValueError(
+                    f"{field} must be provided either directly or through config"
+                )
        return self

    @model_validator(mode="after")
@@ -841,8 +838,8 @@ class Task(BaseModel):
        should_inject = self.allow_crewai_trigger_context

        if should_inject and self.agent:
-            crew = self.agent.crew
-            if crew and not isinstance(crew, str) and crew._inputs:
+            crew = getattr(self.agent, "crew", None)
+            if crew and hasattr(crew, "_inputs") and crew._inputs:
                trigger_payload = crew._inputs.get("crewai_trigger_payload")
                if trigger_payload is not None:
                    description += f"\n\nTrigger Payload: {trigger_payload}"
@@ -855,12 +852,11 @@ class Task(BaseModel):
                    isinstance(self.agent.llm, BaseLLM)
                    and self.agent.llm.supports_multimodal()
                ):
-                    provider: str = self.agent.llm.provider or self.agent.llm.model
-                    api: str | None = (
-                        self.agent.llm.api
-                        if isinstance(self.agent.llm, OpenAICompletion)
-                        else None
+                    provider: str = str(
+                        getattr(self.agent.llm, "provider", None)
+                        or getattr(self.agent.llm, "model", "openai")
                    )
+                    api: str | None = getattr(self.agent.llm, "api", None)
                    supported_types = get_supported_content_types(provider, api)

                def is_auto_injected(content_type: str) -> bool:
--- a/lib/crewai/src/crewai/utilities/pydantic_schema_utils.py
+++ b/lib/crewai/src/crewai/utilities/pydantic_schema_utils.py
@@ -19,18 +19,7 @@ from collections.abc import Callable
 from copy import deepcopy
 import datetime
 import logging
-from typing import (
-    TYPE_CHECKING,
-    Annotated,
-    Any,
-    Final,
-    ForwardRef,
-    Literal,
-    Optional,
-    TypedDict,
-    Union,
-    cast,
-)
+from typing import TYPE_CHECKING, Annotated, Any, Final, Literal, TypedDict, Union, cast
 import uuid

 import jsonref  # type: ignore[import-untyped]
@@ -110,22 +99,15 @@ def resolve_refs(schema: dict[str, Any]) -> dict[str, Any]:
    """
    defs = schema.get("$defs", {})
    schema_copy = deepcopy(schema)
-    expanding: set[str] = set()

    def _resolve(node: Any) -> Any:
        if isinstance(node, dict):
            ref = node.get("$ref")
            if isinstance(ref, str) and ref.startswith("#/$defs/"):
                def_name = ref.replace("#/$defs/", "")
-                if def_name not in defs:
-                    raise KeyError(f"Definition '{def_name}' not found in $defs.")
-                if def_name in expanding:
-                    return {}
-                expanding.add(def_name)
-                try:
+                if def_name in defs:
                    return _resolve(deepcopy(defs[def_name]))
-                finally:
-                    expanding.discard(def_name)
+                raise KeyError(f"Definition '{def_name}' not found in $defs.")
            return {k: _resolve(v) for k, v in node.items()}

        if isinstance(node, list):
@@ -137,11 +119,7 @@ def resolve_refs(schema: dict[str, Any]) -> dict[str, Any]:


 def add_key_in_dict_recursively(
-    d: dict[str, Any],
-    key: str,
-    value: Any,
-    criteria: Callable[[dict[str, Any]], bool],
-    _seen: set[int] | None = None,
+    d: dict[str, Any], key: str, value: Any, criteria: Callable[[dict[str, Any]], bool]
 ) -> dict[str, Any]:
    """Recursively adds a key/value pair to all nested dicts matching `criteria`.

@@ -150,31 +128,22 @@ def add_key_in_dict_recursively(
        key: The key to add.
        value: The value to add.
        criteria: A function that returns True for dicts that should receive the key.
-        _seen: Internal set of visited ``id()``s, used to guard cyclic schemas.

    Returns:
        The modified dictionary.
    """
-    if _seen is None:
-        _seen = set()
    if isinstance(d, dict):
-        if id(d) in _seen:
-            return d
-        _seen.add(id(d))
        if criteria(d) and key not in d:
            d[key] = value
        for v in d.values():
-            add_key_in_dict_recursively(v, key, value, criteria, _seen)
+            add_key_in_dict_recursively(v, key, value, criteria)
    elif isinstance(d, list):
-        if id(d) in _seen:
-            return d
-        _seen.add(id(d))
        for i in d:
-            add_key_in_dict_recursively(i, key, value, criteria, _seen)
+            add_key_in_dict_recursively(i, key, value, criteria)
    return d


-def force_additional_properties_false(d: Any, _seen: set[int] | None = None) -> Any:
+def force_additional_properties_false(d: Any) -> Any:
    """Force additionalProperties=false on all object-type dicts recursively.

    OpenAI strict mode requires all objects to have additionalProperties=false.
@@ -185,17 +154,11 @@ def force_additional_properties_false(d: Any, _seen: set[int] | None = None) ->

    Args:
        d: The dictionary/list to modify.
-        _seen: Internal set of visited ``id()``s, used to guard cyclic schemas.

    Returns:
        The modified dictionary/list.
    """
-    if _seen is None:
-        _seen = set()
    if isinstance(d, dict):
-        if id(d) in _seen:
-            return d
-        _seen.add(id(d))
        if d.get("type") == "object":
            d["additionalProperties"] = False
            if "properties" not in d:
@@ -203,13 +166,10 @@ def force_additional_properties_false(d: Any, _seen: set[int] | None = None) ->
            if "required" not in d:
                d["required"] = []
        for v in d.values():
-            force_additional_properties_false(v, _seen)
+            force_additional_properties_false(v)
    elif isinstance(d, list):
-        if id(d) in _seen:
-            return d
-        _seen.add(id(d))
        for i in d:
-            force_additional_properties_false(i, _seen)
+            force_additional_properties_false(i)
    return d


@@ -223,7 +183,7 @@ OPENAI_SUPPORTED_FORMATS: Final[
 }


-def strip_unsupported_formats(d: Any, _seen: set[int] | None = None) -> Any:
+def strip_unsupported_formats(d: Any) -> Any:
    """Remove format annotations that OpenAI strict mode doesn't support.

    OpenAI only supports: date-time, date, time, duration.
@@ -231,17 +191,11 @@ def strip_unsupported_formats(d: Any, _seen: set[int] | None = None) -> Any:

    Args:
        d: The dictionary/list to modify.
-        _seen: Internal set of visited ``id()``s, used to guard cyclic schemas.

    Returns:
        The modified dictionary/list.
    """
-    if _seen is None:
-        _seen = set()
    if isinstance(d, dict):
-        if id(d) in _seen:
-            return d
-        _seen.add(id(d))
        format_value = d.get("format")
        if (
            isinstance(format_value, str)
@@ -249,17 +203,14 @@ def strip_unsupported_formats(d: Any, _seen: set[int] | None = None) -> Any:
        ):
            del d["format"]
        for v in d.values():
-            strip_unsupported_formats(v, _seen)
+            strip_unsupported_formats(v)
    elif isinstance(d, list):
-        if id(d) in _seen:
-            return d
-        _seen.add(id(d))
        for i in d:
-            strip_unsupported_formats(i, _seen)
+            strip_unsupported_formats(i)
    return d


-def ensure_type_in_schemas(d: Any, _seen: set[int] | None = None) -> Any:
+def ensure_type_in_schemas(d: Any) -> Any:
    """Ensure all schema objects in anyOf/oneOf have a 'type' key.

    OpenAI strict mode requires every schema to have a 'type' key.
@@ -267,17 +218,11 @@ def ensure_type_in_schemas(d: Any, _seen: set[int] | None = None) -> Any:

    Args:
        d: The dictionary/list to modify.
-        _seen: Internal set of visited ``id()``s, used to guard cyclic schemas.

    Returns:
        The modified dictionary/list.
    """
-    if _seen is None:
-        _seen = set()
    if isinstance(d, dict):
-        if id(d) in _seen:
-            return d
-        _seen.add(id(d))
        for key in ("anyOf", "oneOf"):
            if key in d:
                schema_list = d[key]
@@ -285,15 +230,12 @@ def ensure_type_in_schemas(d: Any, _seen: set[int] | None = None) -> Any:
                    if isinstance(schema, dict) and schema == {}:
                        schema_list[i] = {"type": "object"}
                    else:
-                        ensure_type_in_schemas(schema, _seen)
+                        ensure_type_in_schemas(schema)
        for v in d.values():
-            ensure_type_in_schemas(v, _seen)
+            ensure_type_in_schemas(v)
    elif isinstance(d, list):
-        if id(d) in _seen:
-            return d
-        _seen.add(id(d))
        for item in d:
-            ensure_type_in_schemas(item, _seen)
+            ensure_type_in_schemas(item)
    return d


@@ -376,9 +318,7 @@ def add_const_to_oneof_variants(schema: dict[str, Any]) -> dict[str, Any]:
    return _process_oneof(deepcopy(schema))


-def convert_oneof_to_anyof(
-    schema: dict[str, Any], _seen: set[int] | None = None
-) -> dict[str, Any]:
+def convert_oneof_to_anyof(schema: dict[str, Any]) -> dict[str, Any]:
    """Convert oneOf to anyOf for OpenAI compatibility.

    OpenAI's Structured Outputs support anyOf better than oneOf.
@@ -386,37 +326,26 @@ def convert_oneof_to_anyof(

    Args:
        schema: JSON schema dictionary.
-        _seen: Internal set of visited ``id()``s, used to guard cyclic schemas.

    Returns:
        Modified schema with anyOf instead of oneOf.
    """
-    if _seen is None:
-        _seen = set()
    if isinstance(schema, dict):
-        if id(schema) in _seen:
-            return schema
-        _seen.add(id(schema))
        if "oneOf" in schema:
            schema["anyOf"] = schema.pop("oneOf")

        for value in schema.values():
            if isinstance(value, dict):
-                convert_oneof_to_anyof(value, _seen)
+                convert_oneof_to_anyof(value)
            elif isinstance(value, list):
-                if id(value) in _seen:
-                    continue
-                _seen.add(id(value))
                for item in value:
                    if isinstance(item, dict):
-                        convert_oneof_to_anyof(item, _seen)
+                        convert_oneof_to_anyof(item)

    return schema


-def ensure_all_properties_required(
-    schema: dict[str, Any], _seen: set[int] | None = None
-) -> dict[str, Any]:
+def ensure_all_properties_required(schema: dict[str, Any]) -> dict[str, Any]:
    """Ensure all properties are in the required array for OpenAI strict mode.

    OpenAI's strict structured outputs require all properties to be listed
@@ -425,17 +354,11 @@ def ensure_all_properties_required(

    Args:
        schema: JSON schema dictionary.
-        _seen: Internal set of visited ``id()``s, used to guard cyclic schemas.

    Returns:
        Modified schema with all properties marked as required.
    """
-    if _seen is None:
-        _seen = set()
    if isinstance(schema, dict):
-        if id(schema) in _seen:
-            return schema
-        _seen.add(id(schema))
        if schema.get("type") == "object" and "properties" in schema:
            properties = schema["properties"]
            if properties:
@@ -443,21 +366,16 @@ def ensure_all_properties_required(

        for value in schema.values():
            if isinstance(value, dict):
-                ensure_all_properties_required(value, _seen)
+                ensure_all_properties_required(value)
            elif isinstance(value, list):
-                if id(value) in _seen:
-                    continue
-                _seen.add(id(value))
                for item in value:
                    if isinstance(item, dict):
-                        ensure_all_properties_required(item, _seen)
+                        ensure_all_properties_required(item)

    return schema


-def strip_null_from_types(
-    schema: dict[str, Any], _seen: set[int] | None = None
-) -> dict[str, Any]:
+def strip_null_from_types(schema: dict[str, Any]) -> dict[str, Any]:
    """Remove null type from anyOf/type arrays.

    Pydantic generates `T | None` for optional fields, which creates schemas with
@@ -466,17 +384,11 @@ def strip_null_from_types(

    Args:
        schema: JSON schema dictionary.
-        _seen: Internal set of visited ``id()``s, used to guard cyclic schemas.

    Returns:
        Modified schema with null types removed.
    """
-    if _seen is None:
-        _seen = set()
    if isinstance(schema, dict):
-        if id(schema) in _seen:
-            return schema
-        _seen.add(id(schema))
        if "anyOf" in schema:
            any_of = schema["anyOf"]
            non_null = [opt for opt in any_of if opt.get("type") != "null"]
@@ -496,14 +408,11 @@ def strip_null_from_types(

        for value in schema.values():
            if isinstance(value, dict):
-                strip_null_from_types(value, _seen)
+                strip_null_from_types(value)
            elif isinstance(value, list):
-                if id(value) in _seen:
-                    continue
-                _seen.add(id(value))
                for item in value:
                    if isinstance(item, dict):
-                        strip_null_from_types(item, _seen)
+                        strip_null_from_types(item)

    return schema

@@ -542,26 +451,16 @@ _CLAUDE_STRICT_UNSUPPORTED: Final[tuple[str, ...]] = (
 )


-def _strip_keys_recursive(
-    d: Any, keys: tuple[str, ...], _seen: set[int] | None = None
-) -> Any:
+def _strip_keys_recursive(d: Any, keys: tuple[str, ...]) -> Any:
    """Recursively delete a fixed set of keys from a schema."""
-    if _seen is None:
-        _seen = set()
    if isinstance(d, dict):
-        if id(d) in _seen:
-            return d
-        _seen.add(id(d))
        for key in keys:
            d.pop(key, None)
        for v in d.values():
-            _strip_keys_recursive(v, keys, _seen)
+            _strip_keys_recursive(v, keys)
    elif isinstance(d, list):
-        if id(d) in _seen:
-            return d
-        _seen.add(id(d))
        for i in d:
-            _strip_keys_recursive(i, keys, _seen)
+            _strip_keys_recursive(i, keys)
    return d


@@ -820,70 +719,12 @@ def create_model_from_schema(  # type: ignore[no-any-unimported]
    json_schema = force_additional_properties_false(json_schema)
    effective_root = force_additional_properties_false(effective_root)

-    in_progress: dict[int, Any] = {}
-    model = _build_model_from_schema(
-        json_schema,
-        effective_root,
-        model_name=model_name,
-        enrich_descriptions=enrich_descriptions,
-        in_progress=in_progress,
-        __config__=__config__,
-        __base__=__base__,
-        __module__=__module__,
-        __validators__=__validators__,
-        __cls_kwargs__=__cls_kwargs__,
-    )
-
-    types_namespace: dict[str, Any] = {
-        entry.__name__: entry
-        for entry in in_progress.values()
-        if isinstance(entry, type) and issubclass(entry, BaseModel)
-    }
-    for entry in in_progress.values():
-        if (
-            isinstance(entry, type)
-            and issubclass(entry, BaseModel)
-            and not getattr(entry, "__pydantic_complete__", True)
-        ):
-            try:
-                entry.model_rebuild(_types_namespace=types_namespace)
-            except Exception as e:
-                logger.debug("model_rebuild failed for %s: %s", entry.__name__, e)
-    return model
-
-
-def _build_model_from_schema(  # type: ignore[no-any-unimported]
-    json_schema: dict[str, Any],
-    effective_root: dict[str, Any],
-    *,
-    model_name: str | None,
-    enrich_descriptions: bool,
-    in_progress: dict[int, Any],
-    __config__: ConfigDict | None = None,
-    __base__: type[BaseModel] | None = None,
-    __module__: str = __name__,
-    __validators__: dict[str, AnyClassMethod] | None = None,
-    __cls_kwargs__: dict[str, Any] | None = None,
-) -> type[BaseModel]:
-    """Inner builder shared by the public entry point and recursive nested-object creation.
-
-    Preprocessing via ``jsonref.replace_refs`` and the sanitization walkers is
-    run once by the public entry; this helper walks the already-normalized
-    schema and emits Pydantic models. ``in_progress`` maps ``id(schema)`` to
-    the model being built for that schema, so a cyclic ``$ref`` graph
-    degrades to a ``ForwardRef`` back-edge instead of blowing the stack.
-    """
-    original_id = id(json_schema)
    if "allOf" in json_schema:
        json_schema = _merge_all_of_schemas(json_schema["allOf"], effective_root)
+        if "title" not in json_schema and "title" in (root_schema or {}):
+            json_schema["title"] = (root_schema or {}).get("title")

    effective_name = model_name or json_schema.get("title") or "DynamicModel"
-
-    schema_id = id(json_schema)
-    in_progress[original_id] = effective_name
-    if schema_id != original_id:
-        in_progress[schema_id] = effective_name
-
    field_definitions = {
        name: _json_schema_to_pydantic_field(
            name,
@@ -891,14 +732,13 @@ def _build_model_from_schema(  # type: ignore[no-any-unimported]
            json_schema.get("required", []),
            effective_root,
            enrich_descriptions=enrich_descriptions,
-            in_progress=in_progress,
        )
        for name, prop in (json_schema.get("properties", {}) or {}).items()
    }

    effective_config = __config__ or ConfigDict(extra="forbid")

-    model = create_model_base(
+    return create_model_base(
        effective_name,
        __config__=effective_config,
        __base__=__base__,
@@ -907,10 +747,6 @@ def _build_model_from_schema(  # type: ignore[no-any-unimported]
        __cls_kwargs__=__cls_kwargs__,
        **field_definitions,
    )
-    in_progress[original_id] = model
-    if schema_id != original_id:
-        in_progress[schema_id] = model
-    return model


 def _json_schema_to_pydantic_field(
@@ -920,7 +756,6 @@ def _json_schema_to_pydantic_field(
    root_schema: dict[str, Any],
    *,
    enrich_descriptions: bool = False,
-    in_progress: dict[int, Any] | None = None,
 ) -> Any:
    """Convert a JSON schema property to a Pydantic field definition.

@@ -939,7 +774,6 @@ def _json_schema_to_pydantic_field(
        root_schema,
        name_=name.title(),
        enrich_descriptions=enrich_descriptions,
-        in_progress=in_progress,
    )
    is_required = name in required

@@ -999,7 +833,7 @@ def _json_schema_to_pydantic_field(
            field_params["pattern"] = json_schema["pattern"]

    if not is_required:
-        type_ = Optional[type_]  # noqa: UP045 - ForwardRef does not support `|`
+        type_ = type_ | None

    if schema_extra:
        field_params["json_schema_extra"] = schema_extra
@@ -1072,7 +906,6 @@ def _json_schema_to_pydantic_type(
    *,
    name_: str | None = None,
    enrich_descriptions: bool = False,
-    in_progress: dict[int, Any] | None = None,
 ) -> Any:
    """Convert a JSON schema to a Python/Pydantic type.

@@ -1081,23 +914,10 @@ def _json_schema_to_pydantic_type(
        root_schema: The root schema for resolving $ref.
        name_: Optional name for nested models.
        enrich_descriptions: Propagated to nested model creation.
-        in_progress: Map of ``id(schema_dict)`` to the Pydantic model
-            currently being built for that schema, or to a placeholder name
-            as a plain ``str`` while the model is still being constructed.
-            Populated by :func:`_build_model_from_schema`. Enables cycle
-            detection so a self-referential ``$ref`` graph resolves to a
-            :class:`ForwardRef` back-edge rather than recursing forever.

    Returns:
        A Python type corresponding to the JSON schema.
    """
-    if in_progress is not None:
-        cached = in_progress.get(id(json_schema))
-        if isinstance(cached, str):
-            return ForwardRef(cached)
-        if cached is not None:
-            return cached
-
    ref = json_schema.get("$ref")
    if ref:
        ref_schema = _resolve_ref(ref, root_schema)
@@ -1106,7 +926,6 @@ def _json_schema_to_pydantic_type(
            root_schema,
            name_=name_,
            enrich_descriptions=enrich_descriptions,
-            in_progress=in_progress,
        )

    enum_values = json_schema.get("enum")
@@ -1126,7 +945,6 @@ def _json_schema_to_pydantic_type(
                root_schema,
                name_=f"{name_ or 'Union'}Option{i}",
                enrich_descriptions=enrich_descriptions,
-                in_progress=in_progress,
            )
            for i, schema in enumerate(any_of_schemas)
        ]
@@ -1140,15 +958,6 @@ def _json_schema_to_pydantic_type(
                root_schema,
                name_=name_,
                enrich_descriptions=enrich_descriptions,
-                in_progress=in_progress,
-            )
-        if in_progress is not None:
-            return _build_model_from_schema(
-                json_schema,
-                root_schema,
-                model_name=name_,
-                enrich_descriptions=enrich_descriptions,
-                in_progress=in_progress,
            )
        merged = _merge_all_of_schemas(all_of_schemas, root_schema)
        return _json_schema_to_pydantic_type(
@@ -1156,7 +965,6 @@ def _json_schema_to_pydantic_type(
            root_schema,
            name_=name_,
            enrich_descriptions=enrich_descriptions,
-            in_progress=in_progress,
        )

    type_ = json_schema.get("type")
@@ -1177,21 +985,12 @@ def _json_schema_to_pydantic_type(
                root_schema,
                name_=name_,
                enrich_descriptions=enrich_descriptions,
-                in_progress=in_progress,
            )
            return list[item_type]  # type: ignore[valid-type]
        return list
    if type_ == "object":
        properties = json_schema.get("properties")
        if properties:
-            if in_progress is not None:
-                return _build_model_from_schema(
-                    json_schema,
-                    root_schema,
-                    model_name=name_,
-                    enrich_descriptions=enrich_descriptions,
-                    in_progress=in_progress,
-                )
            json_schema_ = json_schema.copy()
            if json_schema_.get("title") is None:
                json_schema_["title"] = name_ or "DynamicModel"
--- a/lib/crewai/tests/a2a/utils/test_delegation.py
+++ b/lib/crewai/tests/a2a/utils/test_delegation.py
@@ -0,0 +1,197 @@
+"""Tests for A2A delegation utilities."""
+
+from __future__ import annotations
+
+from contextlib import asynccontextmanager, ExitStack
+from typing import Any
+from unittest.mock import AsyncMock, MagicMock, patch
+
+import pytest
+
+from a2a.client import Client
+from a2a.types import AgentCapabilities, AgentCard
+
+from crewai.a2a.updates.polling.handler import PollingHandler
+from crewai.a2a.updates.streaming.handler import StreamingHandler
+from crewai.a2a.utils.delegation import get_handler, _aexecute_a2a_delegation_impl
+from crewai.a2a.utils.transport import NegotiatedTransport
+
+
+class TestGetHandler:
+    """Tests for the get_handler helper."""
+
+    def test_returns_streaming_handler_when_config_is_none(self) -> None:
+        assert get_handler(None) is StreamingHandler
+
+    def test_returns_polling_handler_for_polling_config(self) -> None:
+        from crewai.a2a.updates import PollingConfig
+
+        assert get_handler(PollingConfig()) is PollingHandler
+
+
+def _make_agent_card(streaming: bool | None) -> AgentCard:
+    """Build a minimal AgentCard with the given streaming capability."""
+    capabilities = AgentCapabilities(streaming=streaming)
+    return AgentCard(
+        name="test-agent",
+        description="A test agent",
+        url="http://localhost:9999/",
+        version="1.0.0",
+        capabilities=capabilities,
+        defaultInputModes=["text/plain"],
+        defaultOutputModes=["text/plain"],
+        skills=[],
+    )
+
+
+_TASK_RESULT = {"status": "completed", "result": "done", "history": []}
+
+
+def _make_shared_patches(agent_card: AgentCard) -> list:
+    """Return the common patches used across the delegation tests."""
+    mock_client = MagicMock(spec=Client)
+
+    @asynccontextmanager
+    async def _fake_client_ctx(*args: Any, **kwargs: Any):
+        yield mock_client
+
+    negotiated = NegotiatedTransport(
+        transport="JSONRPC",
+        url=agent_card.url,
+        source="server_card",
+    )
+
+    return [
+        patch(
+            "crewai.a2a.utils.delegation._afetch_agent_card_cached",
+            new=AsyncMock(return_value=agent_card),
+        ),
+        patch("crewai.a2a.utils.delegation.validate_auth_against_agent_card"),
+        patch(
+            "crewai.a2a.utils.delegation.validate_required_extensions",
+            return_value=[],
+        ),
+        patch(
+            "crewai.a2a.utils.delegation.negotiate_transport",
+            return_value=negotiated,
+        ),
+        patch(
+            "crewai.a2a.utils.delegation.negotiate_content_types",
+            return_value=MagicMock(output_modes=None),
+        ),
+        patch(
+            "crewai.a2a.utils.delegation._prepare_auth_headers",
+            new=AsyncMock(return_value=({}, None)),
+        ),
+        patch("crewai.a2a.utils.delegation.crewai_event_bus"),
+        patch(
+            "crewai.a2a.utils.delegation._create_a2a_client",
+            side_effect=_fake_client_ctx,
+        ),
+    ]
+
+
+async def _call_impl(agent_card: AgentCard, updates=None) -> None:
+    await _aexecute_a2a_delegation_impl(
+        endpoint="http://localhost:9999/",
+        auth=None,
+        timeout=30,
+        task_description="test task",
+        context=None,
+        context_id=None,
+        task_id=None,
+        reference_task_ids=None,
+        metadata=None,
+        extensions=None,
+        conversation_history=[],
+        is_multiturn=False,
+        turn_number=1,
+        agent_branch=None,
+        agent_id=None,
+        agent_role=None,
+        response_model=None,
+        updates=updates,
+    )
+
+
+class TestStreamingFallback:
+    """Tests that the delegation respects the agent card's streaming capability."""
+
+    @pytest.mark.asyncio
+    async def test_uses_polling_when_agent_card_says_no_streaming(self) -> None:
+        """When streaming=False and updates is None, PollingHandler should be used."""
+        agent_card = _make_agent_card(streaming=False)
+
+        with ExitStack() as stack:
+            for p in _make_shared_patches(agent_card):
+                stack.enter_context(p)
+            mock_polling = stack.enter_context(
+                patch.object(PollingHandler, "execute", new=AsyncMock(return_value=_TASK_RESULT))
+            )
+            mock_streaming = stack.enter_context(
+                patch.object(StreamingHandler, "execute", new=AsyncMock(return_value=_TASK_RESULT))
+            )
+            await _call_impl(agent_card, updates=None)
+
+        mock_polling.assert_called_once()
+        mock_streaming.assert_not_called()
+
+    @pytest.mark.asyncio
+    async def test_uses_streaming_when_agent_card_says_streaming(self) -> None:
+        """When streaming=True and updates is None, StreamingHandler should be used."""
+        agent_card = _make_agent_card(streaming=True)
+
+        with ExitStack() as stack:
+            for p in _make_shared_patches(agent_card):
+                stack.enter_context(p)
+            mock_polling = stack.enter_context(
+                patch.object(PollingHandler, "execute", new=AsyncMock(return_value=_TASK_RESULT))
+            )
+            mock_streaming = stack.enter_context(
+                patch.object(StreamingHandler, "execute", new=AsyncMock(return_value=_TASK_RESULT))
+            )
+            await _call_impl(agent_card, updates=None)
+
+        mock_streaming.assert_called_once()
+        mock_polling.assert_not_called()
+
+    @pytest.mark.asyncio
+    async def test_uses_streaming_when_agent_card_streaming_is_none(self) -> None:
+        """When streaming is unset (None) and updates is None, StreamingHandler should be used."""
+        agent_card = _make_agent_card(streaming=None)
+
+        with ExitStack() as stack:
+            for p in _make_shared_patches(agent_card):
+                stack.enter_context(p)
+            mock_polling = stack.enter_context(
+                patch.object(PollingHandler, "execute", new=AsyncMock(return_value=_TASK_RESULT))
+            )
+            mock_streaming = stack.enter_context(
+                patch.object(StreamingHandler, "execute", new=AsyncMock(return_value=_TASK_RESULT))
+            )
+            await _call_impl(agent_card, updates=None)
+
+        mock_streaming.assert_called_once()
+        mock_polling.assert_not_called()
+
+    @pytest.mark.asyncio
+    async def test_explicit_streaming_config_overrides_agent_card(self) -> None:
+        """Explicitly passing StreamingConfig keeps StreamingHandler even when agent card says no streaming."""
+        from crewai.a2a.updates.streaming.config import StreamingConfig
+
+        agent_card = _make_agent_card(streaming=False)
+
+        with ExitStack() as stack:
+            for p in _make_shared_patches(agent_card):
+                stack.enter_context(p)
+            mock_polling = stack.enter_context(
+                patch.object(PollingHandler, "execute", new=AsyncMock(return_value=_TASK_RESULT))
+            )
+            mock_streaming = stack.enter_context(
+                patch.object(StreamingHandler, "execute", new=AsyncMock(return_value=_TASK_RESULT))
+            )
+            await _call_impl(agent_card, updates=StreamingConfig())
+
+        # explicit config overrides agent card; streaming handler is used
+        mock_streaming.assert_called_once()
+        mock_polling.assert_not_called()
--- a/lib/crewai/tests/agents/test_lite_agent.py
+++ b/lib/crewai/tests/agents/test_lite_agent.py
@@ -1051,7 +1051,7 @@ def test_lite_agent_verbose_false_suppresses_printer_output():
        successful_requests=1,
    )

-    with pytest.warns(FutureWarning):
+    with pytest.warns(DeprecationWarning):
        agent = LiteAgent(
            role="Test Agent",
            goal="Test goal",
--- a/lib/devtools/src/crewai_devtools/init.py
+++ b/lib/devtools/src/crewai_devtools/init.py
@@ -1,3 +1,3 @@
 """CrewAI development tools."""

-__version__ = "1.14.2rc1"
+__version__ = "1.14.2a2"
--- a/lib/devtools/src/crewai_devtools/cli.py
+++ b/lib/devtools/src/crewai_devtools/cli.py
@@ -29,33 +29,6 @@ load_dotenv()
 console = Console()


-def _resume_hint(message: str) -> None:
-    """Print a boxed resume hint after a failure."""
-    console.print()
-    console.print(
-        Panel(
-            message,
-            title="[bold yellow]How to resume[/bold yellow]",
-            border_style="yellow",
-            padding=(1, 2),
-        )
-    )
-
-
-def _print_release_error(e: BaseException) -> None:
-    """Print a release error with stderr if available."""
-    if isinstance(e, KeyboardInterrupt):
-        raise
-    if isinstance(e, SystemExit):
-        return
-    if isinstance(e, subprocess.CalledProcessError):
-        console.print(f"[red]Error running command:[/red] {e}")
-        if e.stderr:
-            console.print(e.stderr)
-    else:
-        console.print(f"[red]Error:[/red] {e}")
-
-
 def run_command(cmd: list[str], cwd: Path | None = None) -> str:
    """Run a shell command and return output.

@@ -291,9 +264,11 @@ def add_docs_version(docs_json_path: Path, version: str) -> bool:
        if not versions:
            continue

+        # Skip if this version already exists for this language
        if any(v.get("version") == version_label for v in versions):
            continue

+        # Find the current default and copy its tabs
        default_version = next(
            (v for v in versions if v.get("default")),
            versions[0],
@@ -305,7 +280,10 @@ def add_docs_version(docs_json_path: Path, version: str) -> bool:
            "tabs": default_version.get("tabs", []),
        }

+        # Remove default flag from old default
        default_version.pop("default", None)
+
+        # Insert new version at the beginning
        versions.insert(0, new_version)
        updated = True

@@ -499,7 +477,7 @@ def _is_crewai_dep(spec: str) -> bool:
    """Return True if *spec* is a ``crewai`` or ``crewai[...]`` dependency."""
    if not spec.startswith("crewai"):
        return False
-    rest = spec[6:]
+    rest = spec[6:]  # after "crewai"
    return len(rest) > 0 and rest[0] in ("[", "=", ">", "<", "~", "!")


@@ -521,6 +499,7 @@ def _pin_crewai_deps(content: str, version: str) -> str:
        deps = doc.get("project", {}).get(key)
        if deps is None:
            continue
+        # optional-dependencies is a table of lists; dependencies is a list
        dep_lists = deps.values() if isinstance(deps, Mapping) else [deps]
        for dep_list in dep_lists:
            for i, dep in enumerate(dep_list):
@@ -659,6 +638,7 @@ def get_github_contributors(commit_range: str) -> list[str]:
        List of GitHub usernames sorted alphabetically.
    """
    try:
+        # Get GitHub token from gh CLI
        try:
            gh_token = run_command(["gh", "auth", "token"])
        except subprocess.CalledProcessError:
@@ -700,6 +680,11 @@ def get_github_contributors(commit_range: str) -> list[str]:
        return []


+# ---------------------------------------------------------------------------
+# Shared workflow helpers
+# ---------------------------------------------------------------------------
+
+
 def _poll_pr_until_merged(
    branch_name: str, label: str, repo: str | None = None
 ) -> None:
@@ -779,6 +764,7 @@ def _update_all_versions(
            "[yellow]Warning:[/yellow] No __version__ attributes found to update"
        )

+    # Update CLI template pyproject.toml files
    templates_dir = lib_dir / "crewai" / "src" / "crewai" / "cli" / "templates"
    if templates_dir.exists():
        if dry_run:
@@ -1177,11 +1163,13 @@ def _repin_crewai_install(run_value: str, version: str) -> str:
    while marker in remainder:
        before, _, after = remainder.partition(marker)
        result.append(before)
+        # after looks like: a2a]==1.14.0" ...
        bracket_end = after.index("]")
        extras = after[:bracket_end]
        rest = after[bracket_end + 1 :]
        if rest.startswith("=="):
-            ver_start = 2
+            # Find end of version — next quote or whitespace
+            ver_start = 2  # len("==")
            ver_end = ver_start
            while ver_end < len(rest) and rest[ver_end] not in ('"', "'", " ", "\n"):
                ver_end += 1
@@ -1343,6 +1331,7 @@ def _release_enterprise(version: str, is_prerelease: bool, dry_run: bool) -> Non
        run_command(["gh", "repo", "clone", enterprise_repo, str(repo_dir)])
        console.print(f"[green]✓[/green] Cloned {enterprise_repo}")

+        # --- bump versions ---
        for rel_dir in _ENTERPRISE_VERSION_DIRS:
            pkg_dir = repo_dir / rel_dir
            if not pkg_dir.exists():
@@ -1372,12 +1361,14 @@ def _release_enterprise(version: str, is_prerelease: bool, dry_run: bool) -> Non
                        f"{pyproject.relative_to(repo_dir)}"
                    )

+        # --- update crewai[tools] pin ---
        enterprise_pyproject = repo_dir / enterprise_dep_path
        if _update_enterprise_crewai_dep(enterprise_pyproject, version):
            console.print(
                f"[green]✓[/green] Updated crewai[tools] dep in {enterprise_dep_path}"
            )

+        # --- update crewai pins in CI workflows ---
        for wf in _update_enterprise_workflows(repo_dir, version):
            console.print(
                f"[green]✓[/green] Updated crewai pin in {wf.relative_to(repo_dir)}"
@@ -1417,6 +1408,7 @@ def _release_enterprise(version: str, is_prerelease: bool, dry_run: bool) -> Non
                time.sleep(_PYPI_POLL_INTERVAL)
        console.print("[green]✓[/green] Workspace synced")

+        # --- branch, commit, push, PR ---
        branch_name = f"feat/bump-version-{version}"
        run_command(["git", "checkout", "-b", branch_name], cwd=repo_dir)
        run_command(["git", "add", "."], cwd=repo_dir)
@@ -1450,6 +1442,7 @@ def _release_enterprise(version: str, is_prerelease: bool, dry_run: bool) -> Non

        _poll_pr_until_merged(branch_name, "enterprise bump PR", repo=enterprise_repo)

+        # --- tag and release ---
        run_command(["git", "checkout", "main"], cwd=repo_dir)
        run_command(["git", "pull"], cwd=repo_dir)

@@ -1491,6 +1484,7 @@ def _trigger_pypi_publish(tag_name: str, wait: bool = False) -> None:
        tag_name: The release tag to publish.
        wait: Block until the workflow run completes.
    """
+    # Capture the latest run ID before triggering so we can detect the new one
    prev_run_id = ""
    if wait:
        try:
@@ -1565,6 +1559,11 @@ def _trigger_pypi_publish(tag_name: str, wait: bool = False) -> None:
        console.print("[green]✓[/green] PyPI publish workflow completed")


+# ---------------------------------------------------------------------------
+# CLI commands
+# ---------------------------------------------------------------------------
+
+
@click.group()
 def cli() -> None:
    """Development tools for version bumping and git automation."""
@@ -1832,80 +1831,62 @@ def release(
        skip_enterprise: Skip the enterprise release phase.
        skip_to_enterprise: Skip phases 1 & 2, run only the enterprise release phase.
    """
-    flags: list[str] = []
-    if no_edit:
-        flags.append("--no-edit")
-    if skip_enterprise:
-        flags.append("--skip-enterprise")
-    flag_suffix = (" " + " ".join(flags)) if flags else ""
-    enterprise_hint = (
-        ""
-        if skip_enterprise
-        else f"\n\nThen release enterprise:\n\n"
-        f"  devtools release {version} --skip-to-enterprise"
-    )
-
-    check_gh_installed()
-
-    if skip_enterprise and skip_to_enterprise:
-        console.print(
-            "[red]Error:[/red] Cannot use both --skip-enterprise "
-            "and --skip-to-enterprise"
-        )
-        sys.exit(1)
-
-    if not skip_enterprise or skip_to_enterprise:
-        missing: list[str] = []
-        if not _ENTERPRISE_REPO:
-            missing.append("ENTERPRISE_REPO")
-        if not _ENTERPRISE_VERSION_DIRS:
-            missing.append("ENTERPRISE_VERSION_DIRS")
-        if not _ENTERPRISE_CREWAI_DEP_PATH:
-            missing.append("ENTERPRISE_CREWAI_DEP_PATH")
-        if missing:
-            console.print(
-                f"[red]Error:[/red] Missing required environment variable(s): "
-                f"{', '.join(missing)}\n"
-                f"Set them or pass --skip-enterprise to skip the enterprise release."
-            )
-            sys.exit(1)
-
-    cwd = Path.cwd()
-    lib_dir = cwd / "lib"
-
-    is_prerelease = _is_prerelease(version)
-
-    if skip_to_enterprise:
-        try:
-            _release_enterprise(version, is_prerelease, dry_run)
-        except BaseException as e:
-            _print_release_error(e)
-            _resume_hint(
-                f"Fix the issue, then re-run:\n\n"
-                f"  devtools release {version} --skip-to-enterprise"
-            )
-            sys.exit(1)
-        console.print(
-            f"\n[green]✓[/green] Enterprise release [bold]{version}[/bold] complete!"
-        )
-        return
-
-    if not dry_run:
-        console.print("Checking git status...")
-        check_git_clean()
-        console.print("[green]✓[/green] Working directory is clean")
-    else:
-        console.print("[dim][DRY RUN][/dim] Would check git status")
-
-    packages = get_packages(lib_dir)
-
-    console.print(f"\nFound {len(packages)} package(s) to update:")
-    for pkg in packages:
-        console.print(f"  - {pkg.name}")
-
-    console.print(f"\n[bold cyan]Phase 1: Bumping versions to {version}[/bold cyan]")
-
    try:
+        check_gh_installed()
+
+        if skip_enterprise and skip_to_enterprise:
+            console.print(
+                "[red]Error:[/red] Cannot use both --skip-enterprise "
+                "and --skip-to-enterprise"
+            )
+            sys.exit(1)
+
+        if not skip_enterprise or skip_to_enterprise:
+            missing: list[str] = []
+            if not _ENTERPRISE_REPO:
+                missing.append("ENTERPRISE_REPO")
+            if not _ENTERPRISE_VERSION_DIRS:
+                missing.append("ENTERPRISE_VERSION_DIRS")
+            if not _ENTERPRISE_CREWAI_DEP_PATH:
+                missing.append("ENTERPRISE_CREWAI_DEP_PATH")
+            if missing:
+                console.print(
+                    f"[red]Error:[/red] Missing required environment variable(s): "
+                    f"{', '.join(missing)}\n"
+                    f"Set them or pass --skip-enterprise to skip the enterprise release."
+                )
+                sys.exit(1)
+
+        cwd = Path.cwd()
+        lib_dir = cwd / "lib"
+
+        is_prerelease = _is_prerelease(version)
+
+        if skip_to_enterprise:
+            _release_enterprise(version, is_prerelease, dry_run)
+            console.print(
+                f"\n[green]✓[/green] Enterprise release [bold]{version}[/bold] complete!"
+            )
+            return
+
+        if not dry_run:
+            console.print("Checking git status...")
+            check_git_clean()
+            console.print("[green]✓[/green] Working directory is clean")
+        else:
+            console.print("[dim][DRY RUN][/dim] Would check git status")
+
+        packages = get_packages(lib_dir)
+
+        console.print(f"\nFound {len(packages)} package(s) to update:")
+        for pkg in packages:
+            console.print(f"  - {pkg.name}")
+
+        # --- Phase 1: Bump versions ---
+        console.print(
+            f"\n[bold cyan]Phase 1: Bumping versions to {version}[/bold cyan]"
+        )
+
        _update_all_versions(cwd, lib_dir, version, packages, dry_run)

        branch_name = f"feat/bump-version-{version}"
@@ -1949,17 +1930,12 @@ def release(
            console.print(
                "[dim][DRY RUN][/dim] Would push branch, create PR, and wait for merge"
            )
-    except BaseException as e:
-        _print_release_error(e)
-        _resume_hint(
-            f"Phase 1 failed. Fix the issue, then re-run:\n\n"
-            f"  devtools release {version}{flag_suffix}"
+
+        # --- Phase 2: Tag and release ---
+        console.print(
+            f"\n[bold cyan]Phase 2: Tagging and releasing {version}[/bold cyan]"
        )
-        sys.exit(1)

-    console.print(f"\n[bold cyan]Phase 2: Tagging and releasing {version}[/bold cyan]")
-
-    try:
        tag_name = version

        if not dry_run:
@@ -1986,57 +1962,22 @@ def release(

        if not dry_run:
            _create_tag_and_release(tag_name, release_notes, is_prerelease)
-    except BaseException as e:
-        _print_release_error(e)
-        _resume_hint(
-            "Phase 2 failed before PyPI publish. The bump PR is already merged.\n"
-            "Fix the issue, then resume with:\n\n"
-            "  devtools tag"
-            f"\n\nAfter tagging, publish to PyPI and update deployment test:\n\n"
-            f"  gh workflow run publish.yml -f release_tag={version}"
-            f"{enterprise_hint}"
-        )
-        sys.exit(1)
-
-    try:
-        if not dry_run:
            _trigger_pypi_publish(tag_name, wait=True)
-    except BaseException as e:
-        _print_release_error(e)
-        _resume_hint(
-            f"Phase 2 failed at PyPI publish. Tag and GitHub release already exist.\n"
-            f"Retry PyPI publish manually:\n\n"
-            f"  gh workflow run publish.yml -f release_tag={version}"
-            f"{enterprise_hint}"
-        )
-        sys.exit(1)
-
-    try:
-        if not dry_run:
            _update_deployment_test_repo(version, is_prerelease)
-    except BaseException as e:
-        _print_release_error(e)
-        _resume_hint(
-            f"Phase 2 failed updating deployment test repo. "
-            f"Tag, release, and PyPI are done.\n"
-            f"Fix the issue and update {_DEPLOYMENT_TEST_REPO} manually."
-            f"{enterprise_hint}"
-        )
-        sys.exit(1)

-    if not skip_enterprise:
-        try:
+        if not skip_enterprise:
            _release_enterprise(version, is_prerelease, dry_run)
-        except BaseException as e:
-            _print_release_error(e)
-            _resume_hint(
-                f"Phase 3 (enterprise) failed. Phases 1 & 2 completed successfully.\n"
-                f"Fix the issue, then resume:\n\n"
-                f"  devtools release {version} --skip-to-enterprise"
-            )
-            sys.exit(1)

-    console.print(f"\n[green]✓[/green] Release [bold]{version}[/bold] complete!")
+        console.print(f"\n[green]✓[/green] Release [bold]{version}[/bold] complete!")
+
+    except subprocess.CalledProcessError as e:
+        console.print(f"[red]Error running command:[/red] {e}")
+        if e.stderr:
+            console.print(e.stderr)
+        sys.exit(1)
+    except Exception as e:
+        console.print(f"[red]Error:[/red] {e}")
+        sys.exit(1)


 cli.add_command(bump)
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -12,7 +12,7 @@ dev = [
    "mypy==1.19.1",
    "pre-commit==4.5.1",
    "bandit==1.9.2",
-    "pytest==9.0.3",
+    "pytest==8.4.2",
    "pytest-asyncio==1.3.0",
    "pytest-subprocess==1.5.3",
    "vcrpy==7.0.0", # pinned, less versions break pytest-recording
@@ -20,7 +20,7 @@ dev = [
    "pytest-randomly==4.0.1",
    "pytest-timeout==2.4.0",
    "pytest-xdist==3.8.0",
-    "pytest-split==0.11.0",
+    "pytest-split==0.10.0",
    "types-requests~=2.31.0.6",
    "types-pyyaml==6.0.*",
    "types-regex==2026.1.15.*",
@@ -162,7 +162,7 @@ info = "Commits must follow Conventional Commits 1.0.0."


 [tool.uv]
-exclude-newer = "1 day"
+exclude-newer = "2026-04-10"  # pinned for CVE-2026-39892; restore to "3 days" after 2026-04-11

 # composio-core pins rich<14 but textual requires rich>=14.
 # onnxruntime 1.24+ dropped Python 3.10 wheels; cap it so qdrant[fastembed] resolves on 3.10.
@@ -170,9 +170,6 @@ exclude-newer = "1 day"
 # langchain-core <1.2.28 has GHSA-926x-3r5x-gfhw (incomplete f-string validation).
 # transformers 4.57.6 has CVE-2026-1839; force 5.4+ (docling 2.84 allows huggingface-hub>=1).
 # cryptography 46.0.6 has CVE-2026-39892; force 46.0.7+.
-# pypdf <6.10.1 has CVE-2026-40260 and GHSA-jj6c-8h6c-hppx; force 6.10.1+.
-# uv <0.11.6 has GHSA-pjjw-68hj-v9mw; force 0.11.6+.
-# python-multipart <0.0.26 has GHSA-mj87-hwqh-73pj; force 0.0.26+.
 override-dependencies = [
    "rich>=13.7.1",
    "onnxruntime<1.24; python_version < '3.11'",
@@ -181,9 +178,6 @@ override-dependencies = [
    "urllib3>=2.6.3",
    "transformers>=5.4.0; python_version >= '3.10'",
    "cryptography>=46.0.7",
-    "pypdf>=6.10.1,<7",
-    "uv>=0.11.6,<1",
-    "python-multipart>=0.0.26,<1",
 ]

 [tool.uv.workspace]
--- a/uv.lock
+++ b/uv.lock
@@ -13,8 +13,7 @@ resolution-markers = [
 ]

 [options]
-exclude-newer = "2026-04-14T20:20:18.36862Z"
-exclude-newer-span = "P1D"
+exclude-newer = "2026-04-10T16:00:00Z"

 [manifest]
 members = [
@@ -28,12 +27,9 @@ overrides = [
    { name = "langchain-core", specifier = ">=1.2.28,<2" },
    { name = "onnxruntime", marker = "python_full_version < '3.11'", specifier = "<1.24" },
    { name = "pillow", specifier = ">=12.1.1" },
-    { name = "pypdf", specifier = ">=6.10.1,<7" },
-    { name = "python-multipart", specifier = ">=0.0.26,<1" },
    { name = "rich", specifier = ">=13.7.1" },
    { name = "transformers", marker = "python_full_version >= '3.10'", specifier = ">=5.4.0" },
    { name = "urllib3", specifier = ">=2.6.3" },
-    { name = "uv", specifier = ">=0.11.6,<1" },
 ]

 [manifest.dependency-groups]
@@ -44,11 +40,11 @@ dev = [
    { name = "mypy", specifier = "==1.19.1" },
    { name = "pip-audit", specifier = "==2.9.0" },
    { name = "pre-commit", specifier = "==4.5.1" },
-    { name = "pytest", specifier = "==9.0.3" },
+    { name = "pytest", specifier = "==8.4.2" },
    { name = "pytest-asyncio", specifier = "==1.3.0" },
    { name = "pytest-randomly", specifier = "==4.0.1" },
    { name = "pytest-recording", specifier = "==0.13.4" },
-    { name = "pytest-split", specifier = "==0.11.0" },
+    { name = "pytest-split", specifier = "==0.10.0" },
    { name = "pytest-subprocess", specifier = "==1.5.3" },
    { name = "pytest-timeout", specifier = "==2.4.0" },
    { name = "pytest-xdist", specifier = "==3.8.0" },
@@ -1356,7 +1352,7 @@ requires-dist = [
    { name = "litellm", marker = "extra == 'litellm'", specifier = "~=1.83.0" },
    { name = "mcp", specifier = "~=1.26.0" },
    { name = "mem0ai", marker = "extra == 'mem0'", specifier = "~=0.1.94" },
-    { name = "openai", specifier = ">=2.0.0,<3" },
+    { name = "openai", specifier = ">=1.83.0,<3" },
    { name = "openpyxl", specifier = "~=3.1.5" },
    { name = "openpyxl", marker = "extra == 'openpyxl'", specifier = "~=3.1.5" },
    { name = "opentelemetry-api", specifier = "~=1.34.0" },
@@ -6728,14 +6724,14 @@ wheels = [

 [[package]]
 name = "pypdf"
-version = "6.10.1"
+version = "6.10.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
    { name = "typing-extensions", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/66/79/f2730c42ec7891a75a2fcea2eb4f356872bcbc671b711418060424796612/pypdf-6.10.1.tar.gz", hash = "sha256:62e6ca7f65aaa28b3d192addb44f97296e4be1748f57ed0f4efb2d4915841880", size = 5315704, upload-time = "2026-04-14T12:55:20.996Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/b8/9f/ca96abf18683ca12602065e4ed2bec9050b672c87d317f1079abc7b6d993/pypdf-6.10.0.tar.gz", hash = "sha256:4c5a48ba258c37024ec2505f7e8fd858525f5502784a2e1c8d415604af29f6ef", size = 5314833, upload-time = "2026-04-10T09:34:57.102Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/f0/04/e3aa7f1f14dbc53429cae34666261eb935d99bd61d24756ab94d7e0309da/pypdf-6.10.1-py3-none-any.whl", hash = "sha256:6331940d3bfe75b7e6601d35db7adabab5fc1d716efaeb384e3c0c3957d033de", size = 335606, upload-time = "2026-04-14T12:55:18.941Z" },
+    { url = "https://files.pythonhosted.org/packages/55/f2/7ebe366f633f30a6ad105f650f44f24f98cb1335c4157d21ae47138b3482/pypdf-6.10.0-py3-none-any.whl", hash = "sha256:90005e959e1596c6e6c84c8b0ad383285b3e17011751cedd17f2ce8fcdfc86de", size = 334459, upload-time = "2026-04-10T09:34:54.966Z" },
 ]

 [[package]]
@@ -6818,7 +6814,7 @@ sdist = { url = "https://files.pythonhosted.org/packages/12/a0/d0638470df605ce26

 [[package]]
 name = "pytest"
-version = "9.0.3"
+version = "8.4.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
    { name = "colorama", marker = "sys_platform == 'win32'" },
@@ -6829,9 +6825,9 @@ dependencies = [
    { name = "pygments" },
    { name = "tomli", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/7d/0d/549bd94f1a0a402dc8cf64563a117c0f3765662e2e668477624baeec44d5/pytest-9.0.3.tar.gz", hash = "sha256:b86ada508af81d19edeb213c681b1d48246c1a91d304c6c81a427674c17eb91c", size = 1572165, upload-time = "2026-04-07T17:16:18.027Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/a3/5c/00a0e072241553e1a7496d638deababa67c5058571567b92a7eaa258397c/pytest-8.4.2.tar.gz", hash = "sha256:86c0d0b93306b961d58d62a4db4879f27fe25513d4b969df351abdddb3c30e01", size = 1519618, upload-time = "2025-09-04T14:34:22.711Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d4/24/a372aaf5c9b7208e7112038812994107bc65a84cd00e0354a88c2c77a617/pytest-9.0.3-py3-none-any.whl", hash = "sha256:2c5efc453d45394fdd706ade797c0a81091eccd1d6e4bccfcd476e2b8e0ab5d9", size = 375249, upload-time = "2026-04-07T17:16:16.13Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/a4/20da314d277121d6534b3a980b29035dcd51e6744bd79075a6ce8fa4eb8d/pytest-8.4.2-py3-none-any.whl", hash = "sha256:872f880de3fc3a5bdc88a11b39c9710c3497a547cfa9320bc3c5e62fbf272e79", size = 365750, upload-time = "2025-09-04T14:34:20.226Z" },
 ]

 [[package]]
@@ -6875,14 +6871,14 @@ wheels = [

 [[package]]
 name = "pytest-split"
-version = "0.11.0"
+version = "0.10.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
    { name = "pytest" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/2f/16/8af4c5f2ceb3640bb1f78dfdf5c184556b10dfe9369feaaad7ff1c13f329/pytest_split-0.11.0.tar.gz", hash = "sha256:8ebdb29cc72cc962e8eb1ec07db1eeb98ab25e215ed8e3216f6b9fc7ce0ec2b5", size = 13421, upload-time = "2026-02-03T09:14:31.469Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/46/d7/e30ba44adf83f15aee3f636daea54efadf735769edc0f0a7d98163f61038/pytest_split-0.10.0.tar.gz", hash = "sha256:adf80ba9fef7be89500d571e705b4f963dfa05038edf35e4925817e6b34ea66f", size = 13903, upload-time = "2024-10-16T15:45:19.783Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ae/a1/d4423657caaa8be9b31e491592b49cebdcfd434d3e74512ce71f6ec39905/pytest_split-0.11.0-py3-none-any.whl", hash = "sha256:899d7c0f5730da91e2daf283860eb73b503259cb416851a65599368849c7f382", size = 11911, upload-time = "2026-02-03T09:14:33.708Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/a7/cad88e9c1109a5c2a320d608daa32e5ee008ccbc766310f54b1cd6b3d69c/pytest_split-0.10.0-py3-none-any.whl", hash = "sha256:466096b086a7147bcd423c6e6c2e57fc62af1c5ea2e256b4ed50fc030fc3dddc", size = 11961, upload-time = "2024-10-16T15:45:18.289Z" },
 ]

 [[package]]
@@ -6989,11 +6985,11 @@ wheels = [

 [[package]]
 name = "python-multipart"
-version = "0.0.26"
+version = "0.0.24"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/88/71/b145a380824a960ebd60e1014256dbb7d2253f2316ff2d73dfd8928ec2c3/python_multipart-0.0.26.tar.gz", hash = "sha256:08fadc45918cd615e26846437f50c5d6d23304da32c341f289a617127b081f17", size = 43501, upload-time = "2026-04-10T14:09:59.473Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/8a/45/e23b5dc14ddb9918ae4a625379506b17b6f8fc56ca1d82db62462f59aea6/python_multipart-0.0.24.tar.gz", hash = "sha256:9574c97e1c026e00bc30340ef7c7d76739512ab4dfd428fec8c330fa6a5cc3c8", size = 37695, upload-time = "2026-04-05T20:49:13.829Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/9a/22/f1925cdda983ab66fc8ec6ec8014b959262747e58bdca26a4e3d1da29d56/python_multipart-0.0.26-py3-none-any.whl", hash = "sha256:c0b169f8c4484c13b0dcf2ef0ec3a4adb255c4b7d18d8e420477d2b1dd03f185", size = 28847, upload-time = "2026-04-10T14:09:58.131Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/73/89930efabd4da63cea44a3f438aeb753d600123570e6d6264e763617a9ce/python_multipart-0.0.24-py3-none-any.whl", hash = "sha256:9b110a98db707df01a53c194f0af075e736a770dc5058089650d70b4a182f950", size = 24420, upload-time = "2026-04-05T20:49:12.555Z" },
 ]

 [[package]]