feat: experimental support for skills.md #7412
Conversation
|
Hi Tibo, I’m excited for this feature to be merged. Are you able to clarify something for me? Does the skill run in its own context or does it pollute the global context? For example say I have a GTM expert skill that navigates a GTM container json (sometimes 100k tokens). When the agent uses the skill to answer a question, it might want to read some or all of the json. Would this then bleed into the main context? |
|
All contributors have signed the CLA ✍️ ✅ |
|
I have read the CLA Document and I hereby sign the CLA |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
tibo-openai
changed the title
| names | ||
| } | ||
|
|
||
| fn merge_project_docs_with_skills( |
There was a problem hiding this comment.
can we get rid of this as well?
| assert_message_ends_with(&request_body["input"][1], "</environment_context>"); | ||
| } | ||
|
|
||
| #[tokio::test(flavor = "multi_thread", worker_threads = 2)] |
There was a problem hiding this comment.
Can we add a separate file skills.rs and follow the pattern that uses test_codex()
This change prototypes support for Skills with the CLI. This is an experimental feature for internal testing.