Repository navigation
Point the skill at config show for project context - #980
Merged
Merged
Conversation
Agents told to check .basecamp/config.json read the local file and miss the repo, global, env and flag layers. config show reports the effective project_id and its source.
Contributor
There was a problem hiding this comment.
🟢 Approved
No active review issues are reported for the change to resolve project context through the CLI’s effective configuration.
No open issues 🎉
What changed in this PR
Updates the Basecamp skill to resolve project context through the CLI’s effective configuration instead of reading only the local config file.
Changes:
- Directs agents to use
basecamp config show --jsonfor the effectiveproject_idand its source. - Adds an evaluation case requiring
config showbeforetodos list, excluding help calls. The model evaluation was not run locally because no API key was available.
[!TIP]
If you aren't ready for review, convert to a draft PR.
Click "Convert to draft" or rungh pr ready --undo.
Click "Ready for review" or rungh pr readyto reengage.
| File | Description |
|---|---|
skills/basecamp/SKILL.md |
Replaces local-file guidance with effective configuration lookup. |
skill-evals/cases/project-context.yml |
Adds a project-context workflow evaluation with command-order checks. |
Review effort: Balanced
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
In a live headless-agent eval, 18 of 22 runs ran
cat .basecamp/config.jsonbecauseskills/basecamp/SKILL.mdsaid to check that file. It is only the local layer: it misses repo, global, env and flag values.basecamp config show --json, whoseproject_idis the effective value with its source.skill-evals/cases/project-context.yml: asked to list todos "in this project", the agent runsconfig showbeforetodos list(help calls do not count). The model eval was not run here (no API key locally).basecamp-doctorandbasecamp-connecthad no such guidance.Summary by cubic
Stops the Basecamp skill from telling agents to read
.basecamp/config.json, which only reflects the local config layer and misses repo, global, env, and flag values. The skill now runsbasecamp config show --jsonto get the effectiveproject_idwith its source.skill-evals/cases/project-context.yml, which requires the agent to runconfig showbeforetodos listwhen asked for todos in this project.Written for commit cbf7a99. Summary will update on new commits.