xianghuijin
5f430076a3
Update documentation for tool_choice auto-reset feature
2025-03-20 21:49:38 +08:00
xianghuijin
eaafad41ca
Revert test file name to match project naming style
2025-03-20 21:28:24 +08:00
xianghuijin
29f39313a3
Rename test file for better clarity
2025-03-20 21:26:59 +08:00
xianghuijin
20271a3591
Fix potential infinite tool call loop by resetting tool_choice after tool execution
2025-03-20 21:22:27 +08:00
Rohan Mehta
d300729e4f
formatting updates to examples doc ( #249 )
2025-03-19 17:26:24 -04:00
James Hills
3de6c69f25
formatting updates to examples doc
2025-03-19 14:23:28 -07:00
Rohan Mehta
b0604f6d8c
add examples section to docs ( #242 )
...
Have gotten feedback that Examples are somewhat buried in the Github
docs. Adding new page after quickstart.
2025-03-19 15:42:41 -04:00
jhills20
266359977d
add examples section to docs
2025-03-19 11:43:09 -07:00
Rohan Mehta
b2a7d1f376
v0.0.5 ( #241 )
2025-03-19 12:35:15 -04:00
Rohan Mehta
a4cfa67160
v0.0.5
2025-03-19 12:32:36 -04:00
Rohan Mehta
66e6cd16e2
Fix breaking changes from openai 1.66.2 ( #232 )
...
Breaking change in 1.66.2
2025-03-19 12:31:59 -04:00
Rohan Mehta
4d6edb9eb9
chore(docs): Added comet opik to tracing AND formatted into a list ( #73 )
...
Comet Opik added support for Agent SDK tracing and should be included.
2025-03-19 12:31:02 -04:00
Rohan Mehta
e7caa14c30
Fix breaking changes from openai 1.66.2
2025-03-19 12:30:45 -04:00
Rohan Mehta
de5710f229
Introduce tool_use_behavior on agents ( #203 )
...
## Context
By default, the outputs of tools are sent to the LLM again. The LLM gets
to read the outputs, and produce a new response. There are cases where
this is not desired:
1. Every tool results in another round trip, and sometimes the output of
the tool is enough.
2. If you force tool use (via model settings `tool_choice=required`),
then the agent will just infinite loop.
This enables you to have different behavior, e.g. use the first tool
output as the final output, or write a custom function to process tool
results and potentially produce an output.
## Test plan
Added new tests and ran existing tests
Also added examples.
Closes #117
2025-03-19 12:27:02 -04:00
Vincent Koc
ef3f1d73b5
Merge branch 'main' into patch-1
2025-03-20 02:57:34 +11:00
Rohan Mehta
435f77cc49
fix reasoning order in guardrails.md ( #234 )
2025-03-18 23:27:33 -04:00
CCM
b642cd4506
fix reasoning order in guardrails.md
2025-03-19 11:10:58 +08:00
Rohan Mehta
c91db09d3c
Introduce tool_use_behavior on agents
2025-03-18 21:55:12 -04:00
Rohan Mehta
8b643537a0
Update tests and docs for strict mode decorator ( #205 )
...
As titled. Test plan: unit tests/docs.
2025-03-18 21:35:00 -04:00
Rohan Mehta
9291c1743c
Add min 95% code coverage ( #229 )
...
Run the tests with `coverage`, and make sure the min coverage is 95%.
2025-03-18 16:50:00 -04:00
Rohan Mehta
d743b4ac4c
Add min 95% code coverage
2025-03-18 16:48:43 -04:00
Dmitry Pimenov
11ce7c9995
chore: adds Maxim AI to tracing processors ( #210 )
2025-03-18 09:10:52 -07:00
Rohan Mehta
9b7cc2436f
[doc] add mlflow tracing integration ( #171 )
...
Hi team! Congrats on the new Agents SDK launch!
This PR adds MLflow to the tracing documentation, as it supports
automatic tracing for OpenAI Agents SDK. It has also been supporting
tracing for
[Swarm](https://mlflow.org/docs/latest/tracing/integrations/swarm ) as
well, so it is fantastic that the new SDK offers built-in mechanism for
3P trace providers🙂

2025-03-18 10:21:54 -04:00
B-Step62
aeb6917e17
[doc] add mlflow tracing integration
...
Signed-off-by: B-Step62 <yuki.watanabe@databricks.com>
2025-03-18 11:29:21 +09:00
Akshay Deo
5b86a4eadc
chore: adds Maxim AI to tracing processors
2025-03-18 07:49:33 +05:30
Dmitry Pimenov
87fb4ee6bc
Docs: update readme text to clarify handoffs ( #208 )
...
Clarifying that the handoff abstraction is implemented via Responses
tool calling
2025-03-17 15:15:13 -07:00
Dmitry Pimenov
e2be8cc604
clarifying that handoffs are a type of tool call under the hood
2025-03-17 15:05:15 -07:00
Rohan Mehta
801cddc28a
Stronger tracing tests with `inline-snapshot` ( #25 )
...
`assert len(spans) == 12` is a very weak assertion. This PR asserts the
exported traces and spans more precisely in a readable tree format. And
when the format of an exported trace/span changes (e.g. a new key is
added to every span), you can use `pytest --inline-snapshot=fix` to
update all relevant tests automatically. See
https://15r10nk.github.io/inline-snapshot/latest/ for more info.
2025-03-17 17:59:04 -04:00
Alex Hall
09279d164e
mypy
2025-03-17 23:56:42 +02:00
Alex Hall
3ff9238dae
Merge branch 'main' of github.com:openai/openai-agents-python into alex/inline-snapshot
2025-03-17 23:55:56 +02:00
Rohan Mehta
7253030b0c
Fix typo on Agent documentation ( #151 )
...
Argument name is not description but handoff_description
2025-03-17 16:51:47 -04:00
Carlos Souza
268ea2f80d
Fix lint
2025-03-17 16:14:28 -04:00
Rohan Mehta
7f1b04e4f3
Fix handoffs example use cases ( #156 )
...
## Changes
- In this handoff example, we need to use `first_agent` with
`random_number_tool`, instead of `second_agent` without that tool, since
the user is asking to generate a random number.
- fixed typo in comments
2025-03-17 15:59:18 -04:00
Rohan Mehta
d5fbf42a9d
fix typo in agent_lifecycle_example.py ( #161 )
2025-03-17 15:59:00 -04:00
Rohan Mehta
2cccc61118
docs: add arize-phoenix to tracing documentation ( #169 )
2025-03-17 15:58:28 -04:00
Rohan Mehta
4bb506e16b
docs: List LangSmith tracing integration ( #197 )
...
LangChain employee/maintainer, happy to make edits to our
docs/integrations if needed.
2025-03-17 15:58:20 -04:00
Rohan Mehta
9876c551bc
fix reasoning order in examples ( #186 )
...
I think we need to put reasoning-related outputs first to encourage CoT
reasoning.
2025-03-17 15:57:52 -04:00
Rohan Mehta
ecf4f92801
Add TracingProcessor export to __init__.py ( #192 )
...
# Summary
This adds the missing TracingProcessor export to __init__.py.
# Behavior
When trying to add a custom tracing processor, the TracingProcessor
importing fails with not found error when trying the example usage
proposed in issue #164
Specifically this line throws the error:
`add_trace_processor(MyTracingProcessor("output"))`
# Expected Behavior
Inspecting the init file, simply the import/export was missing. Adding
these made the example code work for me
# Test plan
Local dev of example code in #164
# Issue number
#164
# Checks
None
2025-03-17 15:39:04 -04:00
Rohan Mehta
8d14695f51
Update tests and docs for strict mode decorator
2025-03-17 15:06:57 -04:00
Dmitry Pimenov
807e95681f
Update tracing docs ( #204 )
...
Update the README to point to the list of tracing processors. The docs
will contain all the external processors, so the readme doesn't get too
long.
2025-03-17 12:02:44 -07:00
Rohan Mehta
18397319e2
Update tracing docs
2025-03-17 14:57:43 -04:00
Rohan Mehta
165366b8e5
Create a utils directory ( #195 )
...
Moves utils files into a separate directory so we don't end up with a
massive utils file
---
[//]: # (BEGIN SAPLING FOOTER)
* #196
* __->__ #195
2025-03-17 14:26:49 -04:00
Rohan Mehta
2ad3bf3fdb
Pretty print result classes ( #196 )
...
These classes print a really long, unreadable output by default. This
trims it to the main useful info (i.e. output, last agent, overview of
run).
---
[//]: # (BEGIN SAPLING FOOTER)
* __->__ #196
* #195
2025-03-17 14:26:43 -04:00
Rohan Mehta
ffdafc107d
Add ZDR callout in tracing docs ( #199 )
...
Add callout for ZDR intracing
2025-03-17 14:26:34 -04:00
James Hills
ff0af156bd
Update tracing.md
...
Add callout for ZDR intracing
2025-03-17 09:13:22 -07:00
Bagatur
f96d352142
docs: List LangSmith tracing integration
2025-03-17 08:47:28 -07:00
Rohan Mehta
417cdc5efe
Pretty print result classes
2025-03-17 11:11:39 -04:00
Vincenzo Domina
f6e59277b1
Add TracingProcessor export to __init__.py
2025-03-17 10:37:43 +01:00
heartkilla
08400fb4aa
fix reasoning order in examples
2025-03-17 14:56:43 +09:00
Rohan Mehta
4719471cc3
utils directory
2025-03-16 18:48:45 -04:00