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.
54
61%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Critical
Do not install without reviewing
Optimize this skill with Tessl
npx tessl skill review --optimize ./.claude/skills/review-fix-loop/SKILL.mdQuality
Discovery
67%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 is strong on specificity and distinctiveness, clearly articulating a unique orchestration role that loops PR review and fix cycles. However, it lacks an explicit 'Use when...' trigger clause and relies on internal skill names rather than natural user language, which limits its effectiveness for skill selection.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user asks to review and fix a pull request end-to-end, or wants an automated review-fix loop until the PR is clean.'
Include natural trigger terms users would say, such as 'pull request', 'clean up my PR', 'review and fix my changes', 'iterate on PR feedback'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: self-review a PR, fix all issues, re-review in a loop until clean, and names the coordinated sub-skills (code-review, address-pr-comments, fix-ci-tests). | 3 / 3 |
Completeness | Clearly answers 'what does this do' (self-review a PR, fix issues, loop until clean, coordinate sub-skills), but lacks an explicit 'Use when...' clause specifying when Claude should select this skill. | 2 / 3 |
Trigger Term Quality | Includes relevant terms like 'PR', 'review', 'code-review', 'CI tests', and 'pr-comments', but misses common natural user phrases like 'pull request', 'check my PR', 'review and fix', or 'clean up PR'. The sub-skill names are hyphenated internal references rather than natural user language. | 2 / 3 |
Distinctiveness Conflict Risk | The description carves out a clear niche as an orchestrator/coordinator skill that loops review-fix cycles, distinguishing it from the individual code-review, address-pr-comments, and fix-ci-tests skills it coordinates. | 3 / 3 |
Total | 10 / 12 Passed |
Implementation
55%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill excels at actionability and workflow clarity — every step has executable commands, explicit gate checks, validation loops, and a rigorous success-streak mechanism. However, it is severely undermined by extreme verbosity: the same GraphQL query is duplicated three times verbatim, security warnings are repeated multiple times, and procedural boilerplate inflates the token count significantly. The monolithic structure with no supporting files compounds the conciseness problem.
Suggestions
Extract the repeated GraphQL pagination query into a single named section or a referenced helper script, and reference it from Steps 2E and 3 instead of duplicating it three times.
Consolidate the security/loop-control warnings into a single prominent section at the top rather than repeating the 'do not read comment bodies' instruction in multiple places throughout the document.
Factor out the summary template, commit message conventions, and the decision matrix into separate referenced files (e.g., TEMPLATES.md, CONVENTIONS.md) to reduce the main skill's length and improve progressive disclosure.
Remove redundant completion check descriptions — define the pattern once (e.g., 'All steps follow the same gate-check pattern: verify prior step completed via TaskList, set current to in_progress') rather than restating it for every step.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~350+ lines. There is massive repetition — the GraphQL pagination query for unresolved threads appears three times in full, the security warning about not reading comment bodies is repeated multiple times, and the gating/completion check boilerplate is restated for every step. Much of this could be factored into a single reference. The execution protocol warnings ('STOP — READ THIS BEFORE DOING ANYTHING ELSE') and anti-skip admonitions are heavily padded. | 1 / 3 |
Actionability | The skill provides fully executable bash commands, specific GraphQL queries, exact CLI invocations (gh pr view, gh pr comment, gh pr checks), concrete commit message formats, and precise sub-skill invocation syntax. Every step has copy-paste ready commands with clear variable substitution patterns. | 3 / 3 |
Workflow Clarity | The multi-step workflow is exceptionally well-sequenced with explicit gate checks between steps, a clear execution order diagram, validation checkpoints at every stage (verify push, confirm CI, confirm unresolved threads), feedback loops (Step 3 failure resets to Step 2), and a SUCCESS_COUNT mechanism requiring 5 consecutive clean passes. The decision matrix in 2E is precise and unambiguous. | 3 / 3 |
Progressive Disclosure | The entire skill is a monolithic wall of text with no references to supporting files. The repeated GraphQL queries, the security warnings, and the detailed sub-step instructions could all be factored into separate referenced documents. There are no bundle files, and the content that could benefit from separation (e.g., the GraphQL query template, the summary format template, the decision matrix) is all inline, making the skill overwhelming to parse. | 1 / 3 |
Total | 8 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- DataDog/rshell
- Commit
00bdc03
- Reviewed
Table of Contents
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.