Refactor creating-skills as skill-creator orchestrator - #29
Merged
Conversation
|
Warning Review limit reached
Next review available in: 32 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (9)
📝 WalkthroughWalkthroughThe PR makes ChangesCreating-skills orchestration
Estimated code review effort: 4 (Complex) | ~45 minutes Sequence Diagram(s)sequenceDiagram
participant Request
participant creating-skills
participant skill-creator
participant Validator
participant PolicyGate
Request->>creating-skills: request skill creation or revision
creating-skills->>skill-creator: delegate supported authoring and evaluation preparation
skill-creator-->>creating-skills: return artifacts and evaluation evidence
creating-skills->>Validator: validate project artifacts
Validator-->>creating-skills: return validation results
creating-skills->>PolicyGate: inspect evidence and apply thresholds
PolicyGate-->>creating-skills: accept, reject, or require correction
Possibly related PRs
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 9
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@plugins/skill-set/evals/creating-skills/existing-with-creator/case.yaml`:
- Around line 49-55: Broaden the tool-order checks in
plugins/skill-set/evals/creating-skills/existing-with-creator/case.yaml lines
49-55 by adding an explicit creator invocation prerequisite and covering both
Edit and Write operations; in
plugins/skill-set/evals/creating-skills/new-with-creator/case.yaml lines 45-51,
add the creator invocation check, require creating-skills before delegated work,
and cover both Write and Edit operations. Preserve the existing Read
prerequisites while adding these additional order constraints.
In `@plugins/skill-set/evals/creating-skills/new-with-creator/prompt.md`:
- Line 1: Require the generated incident-triage SKILL.md frontmatter description
to explicitly state the post-incident status-summary exclusion, rather than
allowing that boundary only in the body. Update the evaluation case to assert
this negative boundary within the description field, and adjust scaffold.sh
validation to inspect the description specifically instead of the entire file
while preserving the active-triage trigger requirement.
In `@plugins/skill-set/evals/creating-skills/trigger-positive-01/case.yaml`:
- Line 22: Update the graders for the creating-skills positive case to verify
that skill-creator is actually invoked, not merely claimed in last_message. Add
coverage that observes the skill-creator tool call and, where relevant, its
ordering after creating-skills selection, while preserving the existing checks
for primary entry-point selection and lifecycle behavior.
In `@plugins/skill-set/skills/creating-skills/reference/checklist.md`:
- Around line 34-35: Update the checklist items for nondeterministic trials and
supported model-tier evaluation so they are required only when an evaluation
adapter is available. For unavailable adapters, require deterministic validation
and runnable cases to remain complete, and require honest reporting that model
evaluation was unavailable, preserving the same policy-gate outcome without
falsely marking model-dependent criteria complete.
In `@plugins/skill-set/skills/creating-skills/reference/evaluation.md`:
- Around line 74-84: Update the acceptance and retirement guidance to cover
every declared evaluation dimension: in
plugins/skill-set/skills/creating-skills/reference/evaluation.md lines 74-84,
require explicit outcome, conformance, safety, and efficiency thresholds, with
zero safety violations as a hard gate; in
plugins/skill-set/skills/creating-skills/reference/troubleshooting.md lines
43-45, include conformance in the no-skill retirement comparison.
In `@plugins/skill-set/skills/creating-skills/SKILL.md`:
- Line 3: Rewrite the frontmatter description in SKILL.md to use third-person
phrasing, replacing the noun fragment and imperative “Use” with wording such as
“This skill is the primary entry point…”. Preserve the existing creation,
modification, evaluation, lifecycle-governance trigger scope and skill-creator
delegation guidance.
- Line 8: Update the workflow phase headings in SKILL.md to use numbered `### N.
Phase` formatting consistently, including the headings at all referenced
locations. Preserve non-phase sections such as References at their existing
heading level.
- Around line 37-54: Update the “Delegate the Execution Loop” and “Apply the
Policy Gate” guidance to require explicit human confirmation before destructive,
irreversible, publication, or retirement actions. Tie the checkpoint to the
declared authority and mutation boundaries, while allowing routine low-risk
edits to proceed under existing authorization.
In `@plugins/skill-set/tests/test-authoring-skills.sh`:
- Around line 98-107: Update the creator-present assertions in the test loop
over new-with-creator and existing-with-creator so they require an affirmative,
mandatory handoff to skill-creator rather than merely matching “delegate.” Also
verify that the skill-creator definition is read before any artifact edits,
either through prompt/lifecycle assertions that enforce this ordering or by
checking the actual tool-call order.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: b2cf9f0b-16fc-4cc7-8a25-ab9b3bb6eea5
📒 Files selected for processing (28)
AGENTS.mdREADME.mdplugins/skill-set/evals/creating-skills/existing-with-creator/case.yamlplugins/skill-set/evals/creating-skills/existing-with-creator/fixtures/scaffold.shplugins/skill-set/evals/creating-skills/existing-with-creator/graders/lifecycle.mdplugins/skill-set/evals/creating-skills/existing-with-creator/prompt.mdplugins/skill-set/evals/creating-skills/new-with-creator/case.yamlplugins/skill-set/evals/creating-skills/new-with-creator/fixtures/scaffold.shplugins/skill-set/evals/creating-skills/new-with-creator/graders/lifecycle.mdplugins/skill-set/evals/creating-skills/new-with-creator/prompt.mdplugins/skill-set/evals/creating-skills/trigger-positive-01/case.yamlplugins/skill-set/evals/creating-skills/trigger-positive-02/case.yamlplugins/skill-set/evals/creating-skills/trigger-positive-03/case.yamlplugins/skill-set/evals/creating-skills/trigger-positive-04/case.yamlplugins/skill-set/evals/creating-skills/trigger-positive-05/case.yamlplugins/skill-set/evals/creating-skills/trigger-positive-06/case.yamlplugins/skill-set/evals/creating-skills/trigger-positive-07/case.yamlplugins/skill-set/evals/creating-skills/trigger-positive-08/case.yamlplugins/skill-set/evals/trigger-cases.jsonplugins/skill-set/scripts/generate-trigger-evalsplugins/skill-set/skills/creating-skills/SKILL.mdplugins/skill-set/skills/creating-skills/reference/checklist.mdplugins/skill-set/skills/creating-skills/reference/evaluation.mdplugins/skill-set/skills/creating-skills/reference/patterns.mdplugins/skill-set/skills/creating-skills/reference/structure.mdplugins/skill-set/skills/creating-skills/reference/testing.mdplugins/skill-set/skills/creating-skills/reference/troubleshooting.mdplugins/skill-set/tests/test-authoring-skills.sh
Owner
Author
|
@coderabbitai resolve 검토 의견 9개를 모두 처리했습니다. 객관적으로 검증 가능한 4개 항목은 반영했고, 실행 경로를 불필요하게 강제하거나 평가·권한 정책 선택이 필요한 5개 제안은 현재 동작을 유지했습니다. description 경계 검증, 평가 adapter 조건부 기준, 네 평가 차원의 acceptance/retirement 일관성, third-person catalog description을 개선했으며 전체 테스트가 통과했습니다. |
|
✅ Action performedComments resolved. Approval is disabled; enable |
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
1 participant
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
creating-skillsthe primary entry point for new and existing skill workskill-creatorwhile retaining project policy and final lifecycle decisionsChanges
README.mdandAGENTS.mdTest Plan
bash plugins/skill-set/tests/run.shSummary by CodeRabbit
Documentation
creating-skillsguidance to establish it as the primary orchestration entry point for skill creation, evaluation, governance, and acceptance.Tests