review-fix-loop

Self-review a PR, fix all issues, and re-review in a loop until clean. Coordinates code-review, address-pr-comments, and fix-ci-tests skills.

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/review-fix-loop/SKILL.md

Quality

Content

81%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is a highly actionable, rigorously sequenced workflow with strong validation and feedback loops, scoring top marks on actionability and workflow clarity. Its main weakness is conciseness — a verbatim-duplicated GraphQL query block and motivational padding could be trimmed — and progressive disclosure is limited by having everything inline in one large file with no reference files.

Suggestions

De-duplicate the GraphQL thread-count query: define it once and reference it from both Step 2E and Step 3, or extract it into a script under scripts/.

Trim motivational lecturing (e.g. 'has caused regressions and wasted iterations in every prior run', 'Anything else is a stall and a bug') — state the rule once and move on.

Move the large final-summary template and/or the GraphQL query into a references/ file and link to it to reduce SKILL.md length and improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient for a complex 30-iteration workflow, but the ~30-line GraphQL pagination block is duplicated verbatim in Step 2E and Step 3, and motivational lecturing ('has caused regressions and wasted iterations in every prior run', 'Anything else is a stall and a bug') is padding Claude does not need.	3 / 5
Actionability	Provides fully executable, copy-paste-ready commands (gh pr view/comment/checks, the complete GraphQL pagination query) and concrete sub-skill invocations (/code-review, /address-pr-comments, /fix-ci-tests) with specific examples like the '[iter 3] Fix...' commit prefix and the exact summary template.	5 / 5
Workflow Clarity	Sequencing is explicit (Step 1 → Step 2 loop → Step 3 → Step 4) with gate checks, per-step completion checks, an explicit decision matrix in 2E, and a feedback loop (Step 3 failure resets SUCCESS_COUNT and returns to Step 2) — matching the checklist-with-validation anchor.	5 / 5
Progressive Disclosure	Well-organized with clear headers, a phase table, and logical section flow for a sequential protocol where inline content is mostly appropriate; the main gap is the duplicated GraphQL block and full summary template that could be factored into a reference file, but no bundle files are present.	4 / 5
	Total	17 / 20 Passed

Description

66%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description clearly conveys what the skill does and names concrete coordinating actions, but omits any explicit 'when to use' trigger guidance and lacks synonyms, which caps completeness and trigger-term quality. It is distinct as a meta-orchestrator over its three sub-skills.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when you need to iteratively self-review a PR and fix all review/CI issues until clean.'

Include natural trigger synonyms users might say, such as 'PR review', 'iterate on review feedback', or 'fix PR comments and CI'.

Replace the generic 'fix all issues' with a more specific phrasing like 'address review comments and fix failing CI tests'.

Dimension	Reasoning	Score
Specificity	Names the PR-review domain and several concrete actions ('Self-review a PR', 'fix all issues', 're-review in a loop until clean') plus coordination of three named skills, but 'fix all issues' is somewhat generic, leaving minor coverage gaps versus the fully comprehensive anchor.	4 / 5
Completeness	The 'what' is clearly stated (self-review, fix, re-review in a loop, coordinate skills), but there is no 'Use when...' clause or equivalent trigger guidance, capping completeness at 3 per the rubric guideline.	3 / 5
Trigger Term Quality	Includes natural terms a user would say ('PR', 'self-review', 'fix', 're-review', 'loop'), giving good keyword coverage, but lacks synonyms, file extensions, or a 'Use when...' trigger phrase, so it stops short of comprehensive.	4 / 5
Distinctiveness Conflict Risk	The iterative PR review-fix loop is a distinct orchestration niche that explicitly coordinates three named sub-skills, with only minor overlap risk against those sub-skills themselves.	4 / 5
	Total	15 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: DataDog/rshell
Path: .claude/skills/review-fix-loop/SKILL.md
Commit: 30a7e9e

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.