Skip to content

Commit 34937f3

Browse files
committed
Add compatibility notes for OpenAI package versions in README and getting started guide
1 parent 4bef7ce commit 34937f3

File tree

2 files changed

+5
-1
lines changed

2 files changed

+5
-1
lines changed

docs/openai_agents/README.md

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -17,4 +17,6 @@ The Durable OpenAI Agents integration combines the familiar OpenAI Agents SDK wi
1717
## Documentation
1818

1919
- [Getting Started](getting-started.md) - Setup and your first durable agent
20-
- [Reference](reference.md) - Complete reference documentation
20+
- [Reference](reference.md) - Complete reference documentation
21+
22+
> Compatibility note: The Durable OpenAI Agents integration has been validated with the OpenAI packages pinned in the sample application's `samples-v2/openai_agents/requirements.txt` (currently `openai==1.107.3` and `openai-agents==0.3.0`). Because the OpenAI packages release frequently and may introduce breaking API or behavioral changes, we recommend pinning to those versions if you face unexpected issues before reporting a bug.

docs/openai_agents/getting-started.md

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -56,6 +56,8 @@ Then install them:
5656
pip install -r requirements.txt
5757
```
5858

59+
> Compatibility note: This integration has been tested with the exact OpenAI package versions defined in the sample app at `samples-v2/openai_agents/requirements.txt` (currently `openai==1.107.3` and `openai-agents==0.3.0`). The OpenAI ecosystem evolves quickly and newer releases may introduce breaking changes. If you encounter unexpected errors, try pinning to those specific versions first to verify whether the issue is due to a version mismatch before filing an issue.
60+
5961
### Configuring Durable Task Scheduler Backend
6062

6163
**Durable Task Scheduler is the preferred backend** for this integration as it provides enhanced performance, better observability, and simplified local development. While not a hard requirement, it's strongly recommended for production workloads.

0 commit comments

Comments
 (0)