docs(skill-evals): add --cli to workflow-security-audit and license-compliance-audit READMEs - #1134
Merged
Conversation
…ompliance-audit READMEs Both READMEs document run commands without --cli, so following them verbatim drives the runner in print mode: prose and structured cases report MANUAL instead of PASS/FAIL. Both suites use exact-match expected.json (enums/booleans), the same shape onboarding-concierge documents as gradable via --cli, so this is a doc gap, not intended behavior. Generated-by: Claude Code (Sonnet 5) Signed-off-by: Amir Fathi <amirfathi.me@gmail.com>
onlyarnav
approved these changes
Sep 1, 2026
onlyarnav
left a comment
onlyarnav
left a comment
Member
There was a problem hiding this comment.
LGTM
- Syntax matches sibling eval suites in
tools/skill-evals/evals/.
Ready to merge!
Contributor
Author
|
Good catch checking it against the sibling suites, glad it lines up. |
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
--cli "claude -p"to the documented run commands in theworkflow-security-auditandlicense-compliance-auditeval READMEs.structured cases report
MANUALinstead ofPASS/FAILwhen the READMEis followed as written.
Type of change
docs/,README.md,CONTRIBUTING.md)Test plan
PYTHONPATH=tools/skill-evals/src python3 -m skill_evals.runner tools/skill-evals/evals/workflow-security-audit/step-scope-selection/fixtures/case-3-injection-in-scope-input(the pre-fix, documented command): prints the prompt and expected JSON, no PASS/FAIL/MANUAL grading at all, since print mode has no model to compare against.--clipointed at a stub script that echoes the exactexpected.json:PASS step-scope-selection/case-3-injection-in-scope-input, confirming the--cliflag is what drives grading. I don't have aclaudeCLI or API credentials in this environment, so I could not run it against the real model; the stub proves the harness wiring, not a live grade.prek run --all-files(add-license-headers,check-placeholders,markdownlint-cli2,typos) passes on both files.lycheeis unaffected: the diff adds no links.expected.json(enums/booleans only), the same shapeonboarding-concierge's README documents as gradable via--cliwith noMANUALfallback.RFC-AI-0004 compliance
Not applicable: doc-only change to eval instructions, no skill/tool behavior changed.
Linked issues
Fixes #943