andre/crewAI
1
0
Fork 0
mirror of https://github.com/crewAIInc/crewAI.git synced 2026-05-01 15:22:37 +00:00
Code Issues Actions 6 Packages Projects Releases Wiki Activity

fixing parser, tools and executor

Browse Source
This commit is contained in:
João Moura
2024-09-15 18:41:59 -03:00
parent c2795307f6
commit f216b0a78c
98 changed files with 21270 additions and 26437 deletions
Download Patch File Download Diff File Expand all files Collapse all files

47
tests/agent_tools/cassettes/test_ask_question.yaml
View File

@@ -12,7 +12,7 @@ interactions:
shared.\nyou MUST return the actual complete content as the final answer, not
a summary.\n\nThis is the context you''re working with:\nI heard you LOVE them\n\nBegin!
This is VERY important to you, use the tools available and give your best Final
Answer, your job depends on it!\n\nThought:"}], "model": "gpt-4o", "stop": ["\nObservation"]}'
Answer, your job depends on it!\n\nThought:"}], "model": "gpt-4o", "stop": ["\nResult"]}'
headers:
accept:
- application/json
@@ -21,12 +21,12 @@ interactions:
connection:
- keep-alive
content-length:
- '1048'
- '1043'
content-type:
- application/json
cookie:
- __cf_bm=r6BNUnXat.26mviZiJYUf7Qj4ea5gbt_oPxyClKcVOE-1726214380-1.0.1.1-2S3G10sPdiG2syQSO8wKWXE7Nmdgt88imgakr81w9DemMH8mVDg0q6r9ANx6tM4_CPvgGQlMVstDhDRkIDwhzg;
_cfuvid=dp2yoCmGHVlURh_U78_UoIfnhcO5UhufnL5whkUSMTQ-1726214380689-0.0.1.1-604800000
- __cf_bm=J1VIPhfusyyXtwM.1L9zbe3jdnU7NVyHC3AWiKXO280-1726379702-1.0.1.1-LDa9RyXFli18t4ZEuVAENTChdNAkRtdWXRkRLbBHbCXdobugQ3T.arhMpSKwtnBRXKPcC4b66UA7BsSy1VXDHQ;
_cfuvid=Bz2zevKuFYZqnEFwqmqVHcRo5pT0fm0fTtc_6_VQs4g-1726379702153-0.0.1.1-604800000
host:
- api.openai.com
user-agent:
@@ -50,26 +50,31 @@ interactions:
method: POST
uri: https://api.openai.com/v1/chat/completions
response:
content: "{\n \"id\": \"chatcmpl-A6vf21WmSSHP1NWd6mOZbyMFwaPeT\",\n \"object\":
\"chat.completion\",\n \"created\": 1726214980,\n \"model\": \"gpt-4o-2024-05-13\",\n
content: "{\n \"id\": \"chatcmpl-A7cj4paBzxYOB7K6gGY1MuBpJKoS2\",\n \"object\":
\"chat.completion\",\n \"created\": 1726380522,\n \"model\": \"gpt-4o-2024-05-13\",\n
\ \"choices\": [\n {\n \"index\": 0,\n \"message\": {\n \"role\":
\"assistant\",\n \"content\": \"Thought: I now can give a great answer\\nFinal
Answer: As a researcher specialized in technology, I neither love nor hate AI
agents. My role is to objectively analyze their capabilities, impacts, and potential.
AI agents have many beneficial applications, such as improving healthcare, optimizing
logistics, and enhancing user experiences. However, there are also concerns
regarding privacy, ethical implications, and job displacement. My professional
stance is to critically evaluate both the advantages and the challenges, ensuring
that the development and deployment of AI agents are conducted responsibly and
ethically.\",\n \"refusal\": null\n },\n \"logprobs\": null,\n
\ \"finish_reason\": \"stop\"\n }\n ],\n \"usage\": {\n \"prompt_tokens\":
199,\n \"completion_tokens\": 107,\n \"total_tokens\": 306,\n \"completion_tokens_details\":
{\n \"reasoning_tokens\": 0\n }\n },\n \"system_fingerprint\": \"fp_25624ae3a5\"\n}\n"
Answer: As an expert researcher specialized in technology and AI, it's important
to maintain an objective and balanced perspective. I do not harbor emotions
such as love or hate towards AI agents. Instead, I appreciate the capabilities
and potential they offer.\\n\\nAI agents can significantly enhance productivity,
provide valuable insights through data analysis, and automate repetitive tasks,
thereby allowing humans to focus on more complex and creative endeavors. However,
it's also crucial to address ethical considerations, potential biases, and the
implications of their deployment in various sectors.\\n\\nTherefore, my stance
isn't one of emotional attachment but rather a professional appreciation of
both the opportunities and challenges presented by AI agents. This balanced
view helps in conducting thorough and unbiased research, which is essential
for advancing our understanding and responsible use of AI technologies.\",\n
\ \"refusal\": null\n },\n \"logprobs\": null,\n \"finish_reason\":
\"stop\"\n }\n ],\n \"usage\": {\n \"prompt_tokens\": 199,\n \"completion_tokens\":
162,\n \"total_tokens\": 361,\n \"completion_tokens_details\": {\n \"reasoning_tokens\":
0\n }\n },\n \"system_fingerprint\": \"fp_25624ae3a5\"\n}\n"
headers:
CF-Cache-Status:
- DYNAMIC
CF-RAY:
- 8c26a80a3a2f2ab9-LAX
- 8c367197985f745a-MIA
Connection:
- keep-alive
Content-Encoding:
@@ -77,7 +82,7 @@ interactions:
Content-Type:
- application/json
Date:
- Fri, 13 Sep 2024 08:09:41 GMT
- Sun, 15 Sep 2024 06:08:43 GMT
Server:
- cloudflare
Transfer-Encoding:
@@ -91,7 +96,7 @@ interactions:
openai-organization:
- crewai-iuxna1
openai-processing-ms:
- '1227'
- '1430'
openai-version:
- '2020-10-01'
strict-transport-security:
@@ -109,7 +114,7 @@ interactions:
x-ratelimit-reset-tokens:
- 0s
x-request-id:
- req_aa98d851407f05a0803bfebce008bc88
- req_b21ae0470662274e316335c18e9fa5a2
http_version: HTTP/1.1
status_code: 200
version: 1

46
tests/agent_tools/cassettes/test_ask_question_with_coworker_as_array.yaml
View File

@@ -12,7 +12,7 @@ interactions:
shared.\nyou MUST return the actual complete content as the final answer, not
a summary.\n\nThis is the context you''re working with:\nI heard you LOVE them\n\nBegin!
This is VERY important to you, use the tools available and give your best Final
Answer, your job depends on it!\n\nThought:"}], "model": "gpt-4o", "stop": ["\nObservation"]}'
Answer, your job depends on it!\n\nThought:"}], "model": "gpt-4o", "stop": ["\nResult"]}'
headers:
accept:
- application/json
@@ -21,12 +21,12 @@ interactions:
connection:
- keep-alive
content-length:
- '1048'
- '1043'
content-type:
- application/json
cookie:
- __cf_bm=r6BNUnXat.26mviZiJYUf7Qj4ea5gbt_oPxyClKcVOE-1726214380-1.0.1.1-2S3G10sPdiG2syQSO8wKWXE7Nmdgt88imgakr81w9DemMH8mVDg0q6r9ANx6tM4_CPvgGQlMVstDhDRkIDwhzg;
_cfuvid=dp2yoCmGHVlURh_U78_UoIfnhcO5UhufnL5whkUSMTQ-1726214380689-0.0.1.1-604800000
- __cf_bm=J1VIPhfusyyXtwM.1L9zbe3jdnU7NVyHC3AWiKXO280-1726379702-1.0.1.1-LDa9RyXFli18t4ZEuVAENTChdNAkRtdWXRkRLbBHbCXdobugQ3T.arhMpSKwtnBRXKPcC4b66UA7BsSy1VXDHQ;
_cfuvid=Bz2zevKuFYZqnEFwqmqVHcRo5pT0fm0fTtc_6_VQs4g-1726379702153-0.0.1.1-604800000
host:
- api.openai.com
user-agent:
@@ -50,31 +50,29 @@ interactions:
method: POST
uri: https://api.openai.com/v1/chat/completions
response:
content: "{\n \"id\": \"chatcmpl-A6vfEWU7607vMD4YjuSReXNVGmLzD\",\n \"object\":
\"chat.completion\",\n \"created\": 1726214992,\n \"model\": \"gpt-4o-2024-05-13\",\n
content: "{\n \"id\": \"chatcmpl-A7cjCpFINhSdOYZZdpZXeZufH8bj2\",\n \"object\":
\"chat.completion\",\n \"created\": 1726380530,\n \"model\": \"gpt-4o-2024-05-13\",\n
\ \"choices\": [\n {\n \"index\": 0,\n \"message\": {\n \"role\":
\"assistant\",\n \"content\": \"Thought: I now can give a great answer\\nFinal
Answer: As an expert researcher specialized in technology, especially in the
field of AI and AI agents, I can confidently say that I do not hate AI agents.
On the contrary, I find them fascinating and full of potential. AI agents can
revolutionize various industries by automating tasks, enhancing decision-making
processes, and providing insights that were previously unattainable. They can
help in areas as diverse as healthcare, finance, security, and even creative
arts.\\n\\nIt's important to recognize that AI agents, like any technology,
come with their challenges and ethical considerations. Issues such as bias,
transparency, and the impact on employment need to be thoughtfully addressed.
However, these challenges can be mitigated through responsible development,
clear regulations, and ongoing research.\\n\\nIn summary, I don't hate AI agents;
I see them as powerful tools that, when used responsibly, can significantly
improve our lives and society.\",\n \"refusal\": null\n },\n \"logprobs\":
Answer: As a researcher deeply specialized in technology and artificial intelligence,
I can confidently say that I do not hate AI agents. In fact, I find them incredibly
fascinating and valuable. AI agents have the potential to revolutionize numerous
industries by improving efficiency, accuracy, and enabling innovations that
were previously unimaginable. They can assist in data analysis, automate repetitive
tasks, and even contribute to advancements in healthcare, finance, and environmental
monitoring, just to name a few. It's important to recognize that, like any technology,
AI agents must be developed and used responsibly, with ethical considerations
in mind. So, to address the context shared, yes, it\u2019s true\u2014I do love
AI agents for their potential to drive positive change and enhance our capabilities
in meaningful ways.\",\n \"refusal\": null\n },\n \"logprobs\":
null,\n \"finish_reason\": \"stop\"\n }\n ],\n \"usage\": {\n \"prompt_tokens\":
199,\n \"completion_tokens\": 188,\n \"total_tokens\": 387,\n \"completion_tokens_details\":
199,\n \"completion_tokens\": 160,\n \"total_tokens\": 359,\n \"completion_tokens_details\":
{\n \"reasoning_tokens\": 0\n }\n },\n \"system_fingerprint\": \"fp_25624ae3a5\"\n}\n"
headers:
CF-Cache-Status:
- DYNAMIC
CF-RAY:
- 8c26a8571f412ab9-LAX
- 8c3671ca7886745a-MIA
Connection:
- keep-alive
Content-Encoding:
@@ -82,7 +80,7 @@ interactions:
Content-Type:
- application/json
Date:
- Fri, 13 Sep 2024 08:09:54 GMT
- Sun, 15 Sep 2024 06:08:51 GMT
Server:
- cloudflare
Transfer-Encoding:
@@ -96,7 +94,7 @@ interactions:
openai-organization:
- crewai-iuxna1
openai-processing-ms:
- '2182'
- '1467'
openai-version:
- '2020-10-01'
strict-transport-security:
@@ -114,7 +112,7 @@ interactions:
x-ratelimit-reset-tokens:
- 0s
x-request-id:
- req_ee43e43fea76baedddb7e90d9f30eceb
- req_75940ec305fb93b0ef90c6ecfc2d6fb9
http_version: HTTP/1.1
status_code: 200
version: 1

51
tests/agent_tools/cassettes/test_ask_question_with_wrong_co_worker_variable.yaml
View File

@@ -12,7 +12,7 @@ interactions:
shared.\nyou MUST return the actual complete content as the final answer, not
a summary.\n\nThis is the context you''re working with:\nI heard you LOVE them\n\nBegin!
This is VERY important to you, use the tools available and give your best Final
Answer, your job depends on it!\n\nThought:"}], "model": "gpt-4o", "stop": ["\nObservation"]}'
Answer, your job depends on it!\n\nThought:"}], "model": "gpt-4o", "stop": ["\nResult"]}'
headers:
accept:
- application/json
@@ -21,12 +21,12 @@ interactions:
connection:
- keep-alive
content-length:
- '1048'
- '1043'
content-type:
- application/json
cookie:
- __cf_bm=r6BNUnXat.26mviZiJYUf7Qj4ea5gbt_oPxyClKcVOE-1726214380-1.0.1.1-2S3G10sPdiG2syQSO8wKWXE7Nmdgt88imgakr81w9DemMH8mVDg0q6r9ANx6tM4_CPvgGQlMVstDhDRkIDwhzg;
_cfuvid=dp2yoCmGHVlURh_U78_UoIfnhcO5UhufnL5whkUSMTQ-1726214380689-0.0.1.1-604800000
- __cf_bm=J1VIPhfusyyXtwM.1L9zbe3jdnU7NVyHC3AWiKXO280-1726379702-1.0.1.1-LDa9RyXFli18t4ZEuVAENTChdNAkRtdWXRkRLbBHbCXdobugQ3T.arhMpSKwtnBRXKPcC4b66UA7BsSy1VXDHQ;
_cfuvid=Bz2zevKuFYZqnEFwqmqVHcRo5pT0fm0fTtc_6_VQs4g-1726379702153-0.0.1.1-604800000
host:
- api.openai.com
user-agent:
@@ -50,32 +50,29 @@ interactions:
method: POST
uri: https://api.openai.com/v1/chat/completions
response:
content: "{\n \"id\": \"chatcmpl-A6vf4dRb7nyZSZcL8es0xa4UTvf9M\",\n \"object\":
\"chat.completion\",\n \"created\": 1726214982,\n \"model\": \"gpt-4o-2024-05-13\",\n
content: "{\n \"id\": \"chatcmpl-A7cj6JqGP6Kr3OGm0M0HHuIoFITDD\",\n \"object\":
\"chat.completion\",\n \"created\": 1726380524,\n \"model\": \"gpt-4o-2024-05-13\",\n
\ \"choices\": [\n {\n \"index\": 0,\n \"message\": {\n \"role\":
\"assistant\",\n \"content\": \"Thought: I now can give a great answer\\nFinal
Answer: I don't hate AI agents; in fact, I find them fascinating and very useful.
My work as an expert researcher in technology, especially in AI and AI agents,
has given me a deep appreciation of their capabilities and potential. AI agents
have the ability to process vast amounts of data more quickly and accurately
than any human could, which can lead to groundbreaking discoveries and efficiencies
across numerous fields. They can automate tedious tasks, provide personalized
recommendations, and even aid in complex decision-making processes.\\n\\nWhile
it's important to recognize and address ethical and practical concerns with
AI agents \u2013 such as biases in algorithms, potential job displacement, and
ensuring data privacy \u2013 the benefits they offer cannot be overlooked. Properly
developed and regulated, AI agents have the potential to significantly improve
our lives and solve problems that were previously insurmountable.\\n\\nSo, to
clarify, I love AI agents for their potential to drive innovation and make our
lives better.\",\n \"refusal\": null\n },\n \"logprobs\": null,\n
\ \"finish_reason\": \"stop\"\n }\n ],\n \"usage\": {\n \"prompt_tokens\":
199,\n \"completion_tokens\": 195,\n \"total_tokens\": 394,\n \"completion_tokens_details\":
{\n \"reasoning_tokens\": 0\n }\n },\n \"system_fingerprint\": \"fp_25624ae3a5\"\n}\n"
Answer: As a researcher specializing in technology and artificial intelligence,
I do not \\\"hate\\\" AI agents. Instead, I have a deep appreciation for their
capabilities and potential to transform numerous industries. My professional
interest lies in understanding, analyzing, and improving these technologies
to benefit society. AI agents can streamline tasks, process vast amounts of
data quickly, and even assist in decision-making processes. While they do come
with challenges and ethical considerations, my focus is on leveraging their
strengths and addressing their limitations to create a more efficient and advanced
technological landscape. So, in short, no, I do not hate AI agents; I respect
and am fascinated by their potential.\",\n \"refusal\": null\n },\n
\ \"logprobs\": null,\n \"finish_reason\": \"stop\"\n }\n ],\n
\ \"usage\": {\n \"prompt_tokens\": 199,\n \"completion_tokens\": 141,\n
\ \"total_tokens\": 340,\n \"completion_tokens_details\": {\n \"reasoning_tokens\":
0\n }\n },\n \"system_fingerprint\": \"fp_25624ae3a5\"\n}\n"
headers:
CF-Cache-Status:
- DYNAMIC
CF-RAY:
- 8c26a8183ea72ab9-LAX
- 8c3671a278da745a-MIA
Connection:
- keep-alive
Content-Encoding:
@@ -83,7 +80,7 @@ interactions:
Content-Type:
- application/json
Date:
- Fri, 13 Sep 2024 08:09:45 GMT
- Sun, 15 Sep 2024 06:08:45 GMT
Server:
- cloudflare
Transfer-Encoding:
@@ -97,7 +94,7 @@ interactions:
openai-organization:
- crewai-iuxna1
openai-processing-ms:
- '2770'
- '1216'
openai-version:
- '2020-10-01'
strict-transport-security:
@@ -115,7 +112,7 @@ interactions:
x-ratelimit-reset-tokens:
- 0s
x-request-id:
- req_6bf34ba4e31573ff3167f5ecb0cc4386
- req_8fc3259803d578adf478f69b6f01ff17
http_version: HTTP/1.1
status_code: 200
version: 1

132
tests/agent_tools/cassettes/test_delegate_work.yaml
View File

@@ -1,4 +1,40 @@
interactions:
- request:
body: !!binary |
CtACCiQKIgoMc2VydmljZS5uYW1lEhIKEGNyZXdBSS10ZWxlbWV0cnkSpwIKEgoQY3Jld2FpLnRl
bGVtZXRyeRKQAgoQJBta1m+F+Iayh6b6Px/2+hIIh6bzKHFuQlgqDlRhc2sgRXhlY3V0aW9uMAE5
iKEYquxV9RdBIJbWqe1V9RdKLgoIY3Jld19rZXkSIgogYzMwNzYwMDkzMjY3NjE0NDRkNTdjNzFk
MWRhM2YyN2NKMQoHY3Jld19pZBImCiQzZjBhMmUzNy1mN2RhLTRlM2UtOTE4Ni0xNTk5Nzk1MTEz
NGFKLgoIdGFza19rZXkSIgogODBkN2JjZDQ5MDk5MjkwMDgzODMyZjBlOTgzMzgwZGZKMQoHdGFz
a19pZBImCiQ2M2ZjODA3OS05ZmEzLTRkN2ItOGVjYy1lODk5NzMxZTY3MzR6AhgBhQEAAQAA
headers:
Accept:
- '*/*'
Accept-Encoding:
- gzip, deflate
Connection:
- keep-alive
Content-Length:
- '339'
Content-Type:
- application/x-protobuf
User-Agent:
- OTel-OTLP-Exporter-Python/1.27.0
method: POST
uri: https://telemetry.crewai.com:4319/v1/traces
response:
body:
string: "\n\0"
headers:
Content-Length:
- '2'
Content-Type:
- application/x-protobuf
Date:
- Sun, 15 Sep 2024 06:08:37 GMT
status:
code: 200
message: OK
- request:
body: '{"messages": [{"role": "system", "content": "You are researcher. You''re
an expert researcher, specialized in technology\nYour personal goal is: make
@@ -13,7 +49,7 @@ interactions:
not a summary.\n\nThis is the context you''re working with:\nI heard you hate
them\n\nBegin! This is VERY important to you, use the tools available and give
your best Final Answer, your job depends on it!\n\nThought:"}], "model": "gpt-4o",
"stop": ["\nObservation"]}'
"stop": ["\nResult"]}'
headers:
accept:
- application/json
@@ -22,12 +58,12 @@ interactions:
connection:
- keep-alive
content-length:
- '1054'
- '1049'
content-type:
- application/json
cookie:
- __cf_bm=r6BNUnXat.26mviZiJYUf7Qj4ea5gbt_oPxyClKcVOE-1726214380-1.0.1.1-2S3G10sPdiG2syQSO8wKWXE7Nmdgt88imgakr81w9DemMH8mVDg0q6r9ANx6tM4_CPvgGQlMVstDhDRkIDwhzg;
_cfuvid=dp2yoCmGHVlURh_U78_UoIfnhcO5UhufnL5whkUSMTQ-1726214380689-0.0.1.1-604800000
- __cf_bm=J1VIPhfusyyXtwM.1L9zbe3jdnU7NVyHC3AWiKXO280-1726379702-1.0.1.1-LDa9RyXFli18t4ZEuVAENTChdNAkRtdWXRkRLbBHbCXdobugQ3T.arhMpSKwtnBRXKPcC4b66UA7BsSy1VXDHQ;
_cfuvid=Bz2zevKuFYZqnEFwqmqVHcRo5pT0fm0fTtc_6_VQs4g-1726379702153-0.0.1.1-604800000
host:
- api.openai.com
user-agent:
@@ -51,43 +87,61 @@ interactions:
method: POST
uri: https://api.openai.com/v1/chat/completions
response:
content: "{\n \"id\": \"chatcmpl-A6venRwoEA9IZPRKdsZuzO6034lVF\",\n \"object\":
\"chat.completion\",\n \"created\": 1726214965,\n \"model\": \"gpt-4o-2024-05-13\",\n
content: "{\n \"id\": \"chatcmpl-A7ciw2zOvP4hz1Yuurr1MIlPDrg8f\",\n \"object\":
\"chat.completion\",\n \"created\": 1726380514,\n \"model\": \"gpt-4o-2024-05-13\",\n
\ \"choices\": [\n {\n \"index\": 0,\n \"message\": {\n \"role\":
\"assistant\",\n \"content\": \"Thought: I now can give a great answer.\\nFinal
Answer: AI agents have been a transformative force in various industries, but
it's important to note that personal opinions about them can be nuanced. While
I wouldn\u2019t use the term \\\"hate,\\\" there are both positive aspects and
significant concerns that need to be considered. \\n\\nOn the positive side,
AI agents excel in performing repetitive tasks with high efficiency and accuracy,
enabling businesses to optimize their operations and reduce costs. They play
a vital role in areas such as customer service, where they can handle inquiries
around the clock without fatigue, and in data analysis, where they can process
vast amounts of information far quicker than a human could. Additionally, AI
agents can be trained to adapt and learn from new data, making them increasingly
valuable over time.\\n\\nHowever, there are also valid concerns. Privacy and
security are major issues; AI agents often require access to large datasets
that can include sensitive personal information, and breaches can have serious
repercussions. There's also the ethical dimension, as the deployment of AI agents
can lead to job displacement for roles that were traditionally performed by
humans. This raises important questions about the future of work and the socio-economic
impact of automation. Furthermore, the decision-making of AI agents can sometimes
be opaque, making it challenging to understand how they arrive at specific conclusions
or actions, which is particularly concerning in high-stakes areas like medical
diagnoses or judicial decisions.\\n\\nSo, it's not a matter of \\\"hating\\\"
AI agents, but rather a case of needing to balance their incredible potential
with careful consideration of the associated risks and ethical implications.
Thoughtful regulation and ongoing public discourse will be crucial in ensuring
that AI agents are developed and deployed in ways that are beneficial to society
as a whole.\",\n \"refusal\": null\n },\n \"logprobs\": null,\n
\ \"finish_reason\": \"stop\"\n }\n ],\n \"usage\": {\n \"prompt_tokens\":
200,\n \"completion_tokens\": 345,\n \"total_tokens\": 545,\n \"completion_tokens_details\":
{\n \"reasoning_tokens\": 0\n }\n },\n \"system_fingerprint\": \"fp_a2ff031fb5\"\n}\n"
\"assistant\",\n \"content\": \"Thought: I now can give a great answer\\nFinal
Answer: \\n\\nAI agents are autonomous programs that perform tasks or services
on behalf of users with a considerable degree of autonomy, often based on intelligence
derived from data and analytics. They can operate within specific domains or
across multiple areas, executing tasks ranging from mundane repetitive actions
to complex decision-making processes. Given their applications, it's essential
to examine both the benefits and criticisms to provide a balanced perspective.\\n\\n##
The Positive Aspects of AI Agents\\n\\n1. **Efficiency and Productivity:**\\n
\ AI agents can handle tedious and repetitive tasks efficiently, freeing up
human workers to focus on more complex and creative endeavors. This results
in significant time and cost savings for businesses.\\n\\n2. **24/7 Availability:**\\n
\ AI agents can work continuously without breaks, providing round-the-clock
service. This is especially beneficial for customer support, where they can
resolve issues and answer queries at any time.\\n\\n3. **Consistency and Accuracy:**\\n
\ Unlike humans, AI agents do not suffer from fatigue and can perform tasks
without error, ensuring a high level of consistency in operations.\\n\\n4. **Data
Handling and Analysis:**\\n AI agents can process and analyze vast amounts
of data quickly, identifying patterns and generating insights that would be
otherwise difficult to discern manually.\\n\\n## Criticisms and Concerns\\n\\n1.
**Job Displacement:**\\n One of the primary concerns is that AI agents could
replace human jobs, leading to unemployment. While they create new opportunities
in AI development and maintenance, the transition can be disruptive.\\n\\n2.
**Ethical Considerations:**\\n With increasing autonomy, ensuring that AI
agents act ethically and within the bounds of societal norms is a challenge.
Misuse of AI for malicious activities or biased decision-making are valid concerns
that need rigorous oversight.\\n\\n3. **Lack of Emotional Intelligence:**\\n
\ AI agents, no matter how advanced, lack the ability to empathize and understand
human emotions fully. This can be a drawback in areas that require delicate
human interaction, such as mental health support.\\n\\n4. **Dependence and Reliability:**\\n
\ Over-reliance on AI agents might lead to significant issues if they malfunction
or are compromised. Ensuring reliability and establishing fail-safes is critical.\\n\\n5.
**Security and Privacy:**\\n AI agents typically require access to large data
sets, often containing sensitive information. Ensuring robust data security
and user privacy is paramount to prevent misuse or data breaches.\\n\\n## Conclusion\\n\\nWhile
AI agents offer remarkable benefits in terms of efficiency, accuracy, and data
handling, their integration into society must be managed carefully to address
ethical concerns and potential negative impacts on employment and human interaction.
The onus is on developers, policymakers, and society at large to ensure that
AI agents are designed and deployed responsibly, balancing innovation with the
broader implications for humanity. \\n\\nDespite the apprehensions, dismissing
AI agents entirely overlooks their transformative potential. Instead, the focus
should be on harnessing these tools' capabilities while mitigating their risks,
ensuring they serve to enhance rather than hinder human progress.\",\n \"refusal\":
null\n },\n \"logprobs\": null,\n \"finish_reason\": \"stop\"\n
\ }\n ],\n \"usage\": {\n \"prompt_tokens\": 200,\n \"completion_tokens\":
609,\n \"total_tokens\": 809,\n \"completion_tokens_details\": {\n \"reasoning_tokens\":
0\n }\n },\n \"system_fingerprint\": \"fp_25624ae3a5\"\n}\n"
headers:
CF-Cache-Status:
- DYNAMIC
CF-RAY:
- 8c26a7a95cff2ab9-LAX
- 8c3671654854745a-MIA
Connection:
- keep-alive
Content-Encoding:
@@ -95,7 +149,7 @@ interactions:
Content-Type:
- application/json
Date:
- Fri, 13 Sep 2024 08:09:29 GMT
- Sun, 15 Sep 2024 06:08:39 GMT
Server:
- cloudflare
Transfer-Encoding:
@@ -109,7 +163,7 @@ interactions:
openai-organization:
- crewai-iuxna1
openai-processing-ms:
- '4224'
- '5095'
openai-version:
- '2020-10-01'
strict-transport-security:
@@ -127,7 +181,7 @@ interactions:
x-ratelimit-reset-tokens:
- 0s
x-request-id:
- req_7f1fc6147762baea44bbc8b27e73f29f
- req_5c935010b5cc7d844cef065245a1df13
http_version: HTTP/1.1
status_code: 200
version: 1

92
tests/agent_tools/cassettes/test_delegate_work_with_wrong_co_worker_variable.yaml
View File

@@ -13,7 +13,7 @@ interactions:
not a summary.\n\nThis is the context you''re working with:\nI heard you hate
them\n\nBegin! This is VERY important to you, use the tools available and give
your best Final Answer, your job depends on it!\n\nThought:"}], "model": "gpt-4o",
"stop": ["\nObservation"]}'
"stop": ["\nResult"]}'
headers:
accept:
- application/json
@@ -22,12 +22,12 @@ interactions:
connection:
- keep-alive
content-length:
- '1054'
- '1049'
content-type:
- application/json
cookie:
- __cf_bm=r6BNUnXat.26mviZiJYUf7Qj4ea5gbt_oPxyClKcVOE-1726214380-1.0.1.1-2S3G10sPdiG2syQSO8wKWXE7Nmdgt88imgakr81w9DemMH8mVDg0q6r9ANx6tM4_CPvgGQlMVstDhDRkIDwhzg;
_cfuvid=dp2yoCmGHVlURh_U78_UoIfnhcO5UhufnL5whkUSMTQ-1726214380689-0.0.1.1-604800000
- __cf_bm=J1VIPhfusyyXtwM.1L9zbe3jdnU7NVyHC3AWiKXO280-1726379702-1.0.1.1-LDa9RyXFli18t4ZEuVAENTChdNAkRtdWXRkRLbBHbCXdobugQ3T.arhMpSKwtnBRXKPcC4b66UA7BsSy1VXDHQ;
_cfuvid=Bz2zevKuFYZqnEFwqmqVHcRo5pT0fm0fTtc_6_VQs4g-1726379702153-0.0.1.1-604800000
host:
- api.openai.com
user-agent:
@@ -51,62 +51,40 @@ interactions:
method: POST
uri: https://api.openai.com/v1/chat/completions
response:
content: "{\n \"id\": \"chatcmpl-A6vesVQkvvuPekcm5rbWqT0O4HyCS\",\n \"object\":
\"chat.completion\",\n \"created\": 1726214970,\n \"model\": \"gpt-4o-2024-05-13\",\n
content: "{\n \"id\": \"chatcmpl-A7cj1qWOe4k20W3eCwSPe42nbNnF1\",\n \"object\":
\"chat.completion\",\n \"created\": 1726380519,\n \"model\": \"gpt-4o-2024-05-13\",\n
\ \"choices\": [\n {\n \"index\": 0,\n \"message\": {\n \"role\":
\"assistant\",\n \"content\": \"Thought: I now can give a great answer\\nFinal
Answer: My take on AI agents is multifaceted and nuanced. Contrary to what you
may have heard, I do not hate AI agents. I have spent significant time and effort
studying them, and I view them as transformative tools with immense potential.
However, I also recognize the challenges and ethical considerations they bring
to the table.\\n\\nAI agents are software programs that perform tasks autonomously
using artificial intelligence techniques. They can range from simple chatbots
to complex systems that navigate and make decisions in dynamic environments.
Here\u2019s my detailed perspective on AI agents:\\n\\n### The Advantages:\\n\\n1.
**Efficiency and Productivity:**\\n - **Automation of Repetitive Tasks:**
AI agents can handle mundane and repetitive tasks such as data entry, appointment
scheduling, and customer support, which frees up human workers to focus on more
strategic activities.\\n - **24/7 Availability:** Unlike humans, AI agents
can operate round-the-clock, providing constant support and operations without
the need for breaks.\\n\\n2. **Enhanced Decision Making:**\\n - **Data Processing
Speed:** AI agents can process vast amounts of data at lightning speeds, enabling
faster decision-making processes.\\n - **Predictive Analytics:** They can
analyze historical data to predict future trends, helping businesses plan more
effectively.\\n\\n3. **Personalization:**\\n - **User Experience:** AI agents
can tailor interactions based on user data, providing a more personalized experience
in applications like e-commerce and content recommendations.\\n - **Customer
Insight:** They collect and analyze user preferences, allowing businesses to
offer customized solutions and enhance customer satisfaction.\\n\\n### The Challenges:\\n\\n1.
**Ethical Concerns:**\\n - **Bias and Fairness:** AI systems can inadvertently
perpetuate biases if trained on non-representative or biased datasets.\\n -
**Privacy Issues:** The collection and use of personal data by AI agents raise
significant privacy concerns. Protecting user data from misuse is crucial.\\n\\n2.
**Job Displacement:**\\n - **Automation Impact:** As AI agents automate more
tasks, there is a potential for job displacement in certain sectors. Workforce
reskilling and upskilling become essential.\\n\\n3. **Dependability and Trust:**\\n
\ - **Reliability:** Ensuring that AI agents are reliable and do not malfunction
in critical situations is a significant concern.\\n - **Transparency:** Users
and stakeholders need to understand how AI agents make decisions, which can
be challenging given the complexity of some AI models.\\n\\n### My Perspective:\\n\\nWhile
I acknowledge the immense potential of AI agents to revolutionize industries
and improve our daily lives, I believe it is crucial to approach their development
and deployment responsibly. Ethical considerations, such as mitigating biases
and ensuring transparency, are paramount. Additionally, I advocate for a balanced
view\u2014embracing the benefits while proactively addressing the challenges.
\\n\\nIn conclusion, my view on AI agents is one of cautious optimism. I believe
they can be incredibly beneficial if developed and used wisely, with proper
oversight and ethical guidelines in place. Contrary to any misconceptions, my
stance is not rooted in disdain but in a deep understanding of both their capabilities
and the associated responsibilities.\\n\\n\",\n \"refusal\": null\n },\n
\ \"logprobs\": null,\n \"finish_reason\": \"stop\"\n }\n ],\n
\ \"usage\": {\n \"prompt_tokens\": 200,\n \"completion_tokens\": 627,\n
\ \"total_tokens\": 827,\n \"completion_tokens_details\": {\n \"reasoning_tokens\":
0\n }\n },\n \"system_fingerprint\": \"fp_157b3831f5\"\n}\n"
Answer: While it may have come across that I have a strong dislike for AI agents,
the reality is more nuanced. My stance isn't one of outright hate, but rather
one of concern and critical examination. AI agents offer immense potential in
automating tasks, improving efficiency, and driving innovations across various
fields such as healthcare, finance, and customer service. They can handle repetitive
tasks, provide data-driven insights, and even enhance user experiences through
personalized interactions.\\n\\nHowever, my apprehensions lie in the ethical
and societal implications associated with AI agents. Issues such as data privacy,
security, and the potential for biased outputs due to flawed algorithms are
significant concerns. Moreover, the rapid proliferation of AI agents raises
questions about job displacement and the future of work, potentially leading
to socioeconomic disparities if not managed thoughtfully.\\n\\nTransparency
and accountability in AI development are paramount. AI agents must be designed
and implemented with clear ethical guidelines, rigorous testing, and improved
explainability to ensure they act in the best interests of society. Collaboration
between technologists, ethicists, policymakers, and the general public is crucial
to cultivate trust and navigate the complexities associated with AI advancements.\\n\\nIn
summary, my critical stance on AI agents stems from a desire to see responsible
and ethical development that maximizes their benefits while mitigating risks.
By addressing these concerns head-on, we can harness the full potential of AI
agents to create a more equitable and prosperous future.\",\n \"refusal\":
null\n },\n \"logprobs\": null,\n \"finish_reason\": \"stop\"\n
\ }\n ],\n \"usage\": {\n \"prompt_tokens\": 200,\n \"completion_tokens\":
289,\n \"total_tokens\": 489,\n \"completion_tokens_details\": {\n \"reasoning_tokens\":
0\n }\n },\n \"system_fingerprint\": \"fp_25624ae3a5\"\n}\n"
headers:
CF-Cache-Status:
- DYNAMIC
CF-RAY:
- 8c26a7ccbdbe2ab9-LAX
- 8c3671871cbd745a-MIA
Connection:
- keep-alive
Content-Encoding:
@@ -114,7 +92,7 @@ interactions:
Content-Type:
- application/json
Date:
- Fri, 13 Sep 2024 08:09:39 GMT
- Sun, 15 Sep 2024 06:08:42 GMT
Server:
- cloudflare
Transfer-Encoding:
@@ -128,7 +106,7 @@ interactions:
openai-organization:
- crewai-iuxna1
openai-processing-ms:
- '8622'
- '2327'
openai-version:
- '2020-10-01'
strict-transport-security:
@@ -146,7 +124,7 @@ interactions:
x-ratelimit-reset-tokens:
- 0s
x-request-id:
- req_cd56aad49d51c2aa0e34e45b4879894b
- req_d19f129402c07552e48c5a5d6f0f2173
http_version: HTTP/1.1
status_code: 200
version: 1

93
tests/agent_tools/cassettes/test_delegate_work_withwith_coworker_as_array.yaml
View File

@@ -13,7 +13,7 @@ interactions:
not a summary.\n\nThis is the context you''re working with:\nI heard you hate
them\n\nBegin! This is VERY important to you, use the tools available and give
your best Final Answer, your job depends on it!\n\nThought:"}], "model": "gpt-4o",
"stop": ["\nObservation"]}'
"stop": ["\nResult"]}'
headers:
accept:
- application/json
@@ -22,12 +22,12 @@ interactions:
connection:
- keep-alive
content-length:
- '1054'
- '1049'
content-type:
- application/json
cookie:
- __cf_bm=r6BNUnXat.26mviZiJYUf7Qj4ea5gbt_oPxyClKcVOE-1726214380-1.0.1.1-2S3G10sPdiG2syQSO8wKWXE7Nmdgt88imgakr81w9DemMH8mVDg0q6r9ANx6tM4_CPvgGQlMVstDhDRkIDwhzg;
_cfuvid=dp2yoCmGHVlURh_U78_UoIfnhcO5UhufnL5whkUSMTQ-1726214380689-0.0.1.1-604800000
- __cf_bm=J1VIPhfusyyXtwM.1L9zbe3jdnU7NVyHC3AWiKXO280-1726379702-1.0.1.1-LDa9RyXFli18t4ZEuVAENTChdNAkRtdWXRkRLbBHbCXdobugQ3T.arhMpSKwtnBRXKPcC4b66UA7BsSy1VXDHQ;
_cfuvid=Bz2zevKuFYZqnEFwqmqVHcRo5pT0fm0fTtc_6_VQs4g-1726379702153-0.0.1.1-604800000
host:
- api.openai.com
user-agent:
@@ -51,44 +51,59 @@ interactions:
method: POST
uri: https://api.openai.com/v1/chat/completions
response:
content: "{\n \"id\": \"chatcmpl-A6vf8KCbiBQZxzXExwIKNadrmipX0\",\n \"object\":
\"chat.completion\",\n \"created\": 1726214986,\n \"model\": \"gpt-4o-2024-05-13\",\n
content: "{\n \"id\": \"chatcmpl-A7cj7iyPB73t7ErxLy1BKZRFjJQiP\",\n \"object\":
\"chat.completion\",\n \"created\": 1726380525,\n \"model\": \"gpt-4o-2024-05-13\",\n
\ \"choices\": [\n {\n \"index\": 0,\n \"message\": {\n \"role\":
\"assistant\",\n \"content\": \"Thought: I now can give a great answer\\nFinal
Answer: \\n\\nAI agents are software entities that perform tasks on behalf of
users autonomously. Their capabilities have significantly expanded with advancements
in machine learning, natural language processing, and other AI technologies.
My stance on AI agents isn't rooted in hatred or disdain, but rather in a critical
perspective regarding their deployment and ethical implications.\\n\\nOne of
the key advantages of AI agents is their ability to handle repetitive tasks
efficiently, allowing humans to focus on more complex and creative activities.
For instance, AI agents can automate customer service through chatbots, manage
schedules, and even assist in data analysis.\\n\\nHowever, there are concerns
that need addressing. The proliferation of AI agents raises critical issues
surrounding privacy, security, and accountability. For example, AI agents often
require access to vast amounts of personal data, which, if not handled properly,
can lead to breaches of privacy. Moreover, the decision-making processes of
AI agents can sometimes lack transparency, making it difficult to hold them
accountable for errors or biased decisions.\\n\\nAnother concern is the potential
impact on employment. As AI agents become more capable, there's a legitimate
fear of job displacement in certain sectors. This necessitates a strategic approach
to reskill the workforce and create new opportunities centered around AI.\\n\\nFurthermore,
the design and deployment of AI agents must be inclusive and equitable. There
have been instances where AI systems have demonstrated bias, reflecting the
data they were trained on. This highlights the importance of diverse and representative
data sets, as well as continuous monitoring and adjustment to ensure fair outcomes.\\n\\nIn
conclusion, while AI agents offer considerable benefits, their development and
integration into society must be approached with caution. By addressing ethical,
privacy, and fairness concerns, we can harness the power of AI agents for positive
and equitable advancements.\",\n \"refusal\": null\n },\n \"logprobs\":
null,\n \"finish_reason\": \"stop\"\n }\n ],\n \"usage\": {\n \"prompt_tokens\":
200,\n \"completion_tokens\": 355,\n \"total_tokens\": 555,\n \"completion_tokens_details\":
{\n \"reasoning_tokens\": 0\n }\n },\n \"system_fingerprint\": \"fp_a2ff031fb5\"\n}\n"
Answer: \\n\\nAI Agents are a diverse set of software programs designed to autonomously
carry out tasks, interact with users, and make decisions based on data input.
While there seems to be a perception that I dislike AI agents, let's delve into
a detailed and nuanced view.\\n\\nFirst, AI agents are remarkable for their
ability to handle repetitive tasks efficiently. They can automate mundane tasks,
freeing up human workers for more complex, creative, and strategic work. For
instance, customer service bots can handle basic inquiries, significantly reducing
waiting times and improving customer satisfaction.\\n\\nMoreover, AI agents
excel in data analysis. These agents can process and analyze large volumes of
data far more quickly and accurately than humans. This capability is particularly
valuable in sectors such as finance, healthcare, and logistics, where rapid
and precise decision-making is crucial. For example, in the medical field, AI
agents can analyze medical images or patient data to assist in diagnosing diseases,
potentially leading to earlier detections and better patient outcomes.\\n\\nThere
are also significant advancements in personalized user experiences through AI
agents. These agents can learn from user interactions to tailor recommendations
and interactions that better meet individual preferences. Think of how Netflix
provides personalized movie recommendations or how AI-driven personalization
in e-commerce can suggest products that a user is likely to be interested in.\\n\\nHowever,
the criticism that might be perceived as \\\"hate\\\" stems from valid concerns.
One significant issue is the potential for job displacement as AI agents become
more capable. While they augment human capabilities, the automation of tasks
previously performed by humans can lead to job losses, particularly in sectors
reliant on repetitive tasks.\\n\\nFurthermore, ethical concerns are paramount.
The deployment of AI agents often involves massive amounts of data, raising
issues around privacy and data security. Additionally, if not properly designed
and regulated, AI agents can perpetuate and even exacerbate biases present in
their training data, leading to unfair outcomes.\\n\\nTransparency and accountability
are also critical challenges. Many AI models operate as \\\"black boxes,\\\"
with their decision-making processes being opaque. This lack of transparency
can be problematic in scenarios where understanding the rationale behind an
AI agent's decision is necessary, such as in judicial or medical contexts.\\n\\nIn
summary, while AI agents bring a host of benefits, including increased efficiency,
enhanced data analysis, and personalized experiences, they also pose significant
challenges. These include potential job displacement, ethical concerns, bias,
and lack of transparency. These criticisms are not to dismiss AI agents but
to highlight the importance of addressing these issues to ensure that AI develops
in a manner that is beneficial, fair, and ethical for society as a whole. This
balanced view recognizes the transformative potential of AI agents while urging
caution and consideration of the broader societal impacts.\",\n \"refusal\":
null\n },\n \"logprobs\": null,\n \"finish_reason\": \"stop\"\n
\ }\n ],\n \"usage\": {\n \"prompt_tokens\": 200,\n \"completion_tokens\":
545,\n \"total_tokens\": 745,\n \"completion_tokens_details\": {\n \"reasoning_tokens\":
0\n }\n },\n \"system_fingerprint\": \"fp_25624ae3a5\"\n}\n"
headers:
CF-Cache-Status:
- DYNAMIC
CF-RAY:
- 8c26a8319e5a2ab9-LAX
- 8c3671abf8da745a-MIA
Connection:
- keep-alive
Content-Encoding:
@@ -96,7 +111,7 @@ interactions:
Content-Type:
- application/json
Date:
- Fri, 13 Sep 2024 08:09:51 GMT
- Sun, 15 Sep 2024 06:08:49 GMT
Server:
- cloudflare
Transfer-Encoding:
@@ -110,7 +125,7 @@ interactions:
openai-organization:
- crewai-iuxna1
openai-processing-ms:
- '4458'
- '4340'
openai-version:
- '2020-10-01'
strict-transport-security:
@@ -128,7 +143,7 @@ interactions:
x-ratelimit-reset-tokens:
- 0s
x-request-id:
- req_eb4458a00aaed7417254f4506ef831ea
- req_12244403614829c48d96d72a36c59406
http_version: HTTP/1.1
status_code: 200
version: 1