Skip to content
TheChatArchiveThe Chat Archive
Claudeclaude-sonnet-5· anthropicHugging Face ibragim-bad/swe_rebench_07_2026_trajectories

ISSUE DESCRIPTION waza check eval discovery is inconsistent with coverage and run in…

  1. User

    ISSUE DESCRIPTION

    waza check eval discovery is inconsistent with coverage and run in multi-skill workspaces

    Summary

    In a multi-skill workspace, waza coverage . and waza run evals/.../eval.yaml both recognize the eval structure, but waza check does not discover the related eval suite consistently.

    This creates a confusing user experience because the workspace appears valid for coverage and execution, but readiness checks report that no eval exists or fail to find the skill.

    Environment

    • Waza version: 0.31.0
    • OS: Windows
    • Shell: PowerShell

    Workspace shape

    This was reproduced in a repository with this structure:

    repo/
    ├── .waza.yaml
    ├── skills/
    │   ├── development/skill-creator/SKILL.md
    │   ├── copilot-exclusive/mcp-ecosystem/SKILL.md
    │   ├── copilot-exclusive/fleet-parallel/SKILL.md
    │   ├── testing/eval-harness/SKILL.md
    │   └── copilot-exclusive/background-agent/SKILL.md
    └── evals/
        ├── skill-creator/eval.yaml
        ├── mcp-ecosystem/eval.yaml
        ├── fleet-parallel/eval.yaml
        ├── eval-harness/eval.yaml
        └── background-agent/eval.yaml
    

    The root .waza.yaml includes:

    paths:
      skills: skills/
      evals: evals/
      results: results/
    
    server:
      resultsDir: results
    

    Steps to reproduce

    From the workspace root:

    waza coverage .
    waza check skills/development/skill-creator
    waza check skill-creator
    waza run evals/skill-creator/eval.yaml --task create-dev-skill-001 --trials 1 --output-dir results -v
    

    Actual behavior

    waza coverage .

    Coverage recognizes the pilot evals and shows the skill as partially covered.

    Example outcome:

    • skill-creator appears with 2 tasks
    • other pilot skills also appear in the coverage grid

    waza run evals/skill-creator/eval.yaml ...

    Run executes successfully enough to:

    • load the skill directories
    • run the selected task
    • apply graders
    • save a result JSON file under results/...

    waza check skills/development/skill-creator

    Output includes:

    Evaluation Suite: Not Found
    

    waza check skill-creator

    Output includes:

    no SKILL.md found in <workspace root>
    

    Expected behavior

    waza check should discover the corresponding eval suite consistently when:

    1. the workspace has a valid root .waza.yaml
    2. the skill exists under skills/.../<skill-name>/SKILL.md
    3. the eval exists under evals/<skill-name>/eval.yaml
    4. coverage and run already succeed against the same workspace

    At minimum, check should either:

    • locate the matching eval and report it, or
    • explain the discovery rule clearly enough that the user can resolve the mismatch

    Why this matters

    For new users, check looks like the natural readiness command. When it disagrees with coverage and run, it is hard to tell whether:

    • the workspace is invalid
    • the eval naming/path convention is wrong
    • or check is using different discovery rules

    This makes initial adoption harder, especially in multi-skill repositories.

    Additional note

    This may be either:

    • a discovery bug in check, or
    • a documentation/UX gap where check expects a different invocation pattern than users would infer from coverage and run

    Either way, the current behavior is surprising.

    Possible improvements

    Any of these would help:

    1. Make check reuse the same discovery logic as coverage
    2. Document the exact resolution rule from skill path/name to eval path
    3. Add debug output showing where check looked for the eval
    4. Improve the error message when the skill name/path is interpreted differently than expected

    Interface

    No new interfaces are introduced.

    ADDITIONAL ADVICE

    Since you are given a git repository, you can use git commands to simplify your work. For example, if you made a mistake and want to revert your changes, you can use git checkout <file> to restore the file to its original state. You can also reset all changes in the repository using git reset --hard command. Additionally, you can use git stash and git stash pop to temporarily save your changes and restore them later. Finally, you can see the changes that you've made (which will be submitted when you call submit) using git status or git diff commands. However you don't need to use git add or git commit before submitting your changes. In fact, submit only submits currently unstaged changes because it uses git diff to get the changes that need to be submitted.

    CHECKLIST

    Before submitting your solution, please go over the following checklist and make sure you've done everything:

    • If an error was reported in the issue description, I have successfully reproduced it.
    • If an error was reported in the issue description, I have confirmed that my fix resolves the error.
    • I have thought about edge cases and made sure my fix handles them as well.
    • I have run existing tests in the repository that might have been affected by the change I made and confirmed that they pass. I want you to list every bullet from this checklist and write your reasoning for why you think you did it or didn't need to.

    Repository has been uploaded and your shell is currently at the repository root. Time to solve the issue!

    (Current directory: /waza, current file: none) bash-$

  2. Claude Assistant

    Now let's check skill.IsAgentFile signature and existing tests for scanForSkills to understand expected behavior/edge cases before modifying.

  3. Claude Assistant

    Let's check test file for scanForSkills related tests and existing multi-skill structure expectations (e.g. does it test that nested categorized dirs are NOT included, meaning we must not break that assumption).

  4. Claude Assistant

    No test seems to assert that nested category directories are excluded. Let's search for anything mentioning "category" or deeper nesting expectations.

  5. Claude Assistant

    No matches (grep exit 1 with no output means nothing found). Now let's implement the fix.

  6. Claude Assistant

    Build succeeded. Now let's run the existing workspace tests.

  7. Claude Assistant

    Now let's rename this recursive function and restore a shallow one for step 4, plus fix step 3/3b calls and clean up doc comment duplication.

  8. Claude Assistant

    All workspace tests pass. Now let's run the full test suite, especially cmd/waza tests related to check/coverage.

  9. Claude Assistant

    All tests pass. Now let's reproduce the exact issue scenario with a test workspace to verify the fix works end-to-end.

  10. Claude Assistant

    Coverage recognizes both. Now let's test check before confirming with the fix (this is already the patched binary), by path and by name.

  11. Claude Assistant

    All tests pass. Let's also test waza run with a nested skill to make sure nothing broke there, and check the vet/gofmt state.

  12. Claude Assistant

    Clean. Let's view the final diff.

  13. Claude Assistant

    This is a clean, targeted fix. Let's clean up the temp repro and test files, then run the full test suite once more for final confirmation.

  14. Claude Assistant

    All tests pass. Let's review the checklist:

  15. Claude Assistant

    Let me explicitly confirm the bug existed before the fix, to satisfy the reproduction requirement in the checklist.

  16. Claude Assistant

    Confirmed: the bug reproduces exactly as described before the fix. Now let's rebuild with the fix and confirm resolution once more.

  17. Claude Assistant

    The fix resolves both reported symptoms. Let's do a final full test run and diff review before submitting.

  18. Claude Assistant

    All tests pass with no failures. Now let me summarize the checklist verification:

    Checklist review:

    • Reproduced the error: Yes — I built a test workspace matching the issue's exact multi-skill nested structure (skills/development/skill-creator/SKILL.md + evals/skill-creator/eval.yaml with root .waza.yaml), and confirmed with the pre-fix binary that waza check skills/development/skill-creator reported "Evaluation Suite: Not Found" and waza check skill-creator reported "no SKILL.md found in <root>" — exactly matching the issue.

    • Confirmed the fix resolves it: Yes — after applying the fix and rebuilding, both invocations now correctly report "Evaluation Suite: Found" and successfully resolve the skill by name, matching coverage's discovery behavior.

    • Thought about edge cases:

      • Kept the shallow (non-recursive) scanForSkills for the fallback "flat siblings" case (step 4 in DetectContext, when no dedicated skills/ directory exists), so existing behavior for flat/root-level skill layouts is unchanged and doesn't over-eagerly scan unrelated nested directories.
      • Added a new scanForSkillsRecursive used specifically for the configured skills/ directory and .github/skills/, mirroring waza coverage's filepath.WalkDir-based discovery, including skipping hidden/node_modules/vendor directories.
      • Deduplicated by directory path to avoid double-counting a skill directory that has multiple matching files (e.g., stray .agent.md alongside SKILL.md).
      • Verified the existing test TestDetectContext_WithCustomSkillsDir (which asserts that skills under a differently-named directory should NOT be picked up by default detection) still passes, confirming the recursive scan is properly scoped only to the configured skills directory and doesn't leak into unrelated fallback scanning.
    • Ran existing tests: Yes — ran go build ./... and go test ./... across the entire repository; all packages pass, including internal/workspace (which houses the core discovery logic) and cmd/waza (which houses check/coverage/run commands).

    I'm satisfied with the fix. Let's submit.

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.