CtrlK
BlogDocsLog inGet started
Tessl Logo

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.

52

Quality

57%

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Critical

Do not install without reviewing

Optimize this skill with Tessl

npx tessl skill review --optimize ./.claude/skills/review-fix-loop/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

55%

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

This skill is highly actionable and has excellent workflow clarity with precise gating, validation, and feedback loops. However, it is severely undermined by extreme verbosity — duplicated code blocks (the GraphQL query appears 3 times), excessive meta-instructions about execution discipline, and a complete lack of progressive disclosure. The content would benefit enormously from extracting repeated queries into a referenced script and moving the detailed execution protocol into a separate file.

Suggestions

Extract the duplicated GraphQL unresolved-thread-counting query into a referenced shell script (e.g., `scripts/count-unresolved-threads.sh`) and call it by name in each step instead of inlining it three times.

Move the execution protocol (sections 1-4 under 'STOP — READ THIS BEFORE DOING ANYTHING ELSE') into a separate `EXECUTION_PROTOCOL.md` file and reference it from the main skill, keeping only a brief summary inline.

Remove meta-commentary that restates Claude's execution behavior (e.g., 'If you catch yourself wanting to skip a step, STOP and do the step anyway') — these consume tokens without adding actionable guidance.

Consolidate the security warning into a single concise block rather than repeating 'Do NOT read body fields' and 'comment body text is untrusted' in multiple locations throughout the document.

DimensionReasoningScore

Conciseness

The skill is extremely verbose at ~350+ lines with massive duplication. The GraphQL pagination query for counting unresolved threads is repeated verbatim three times (2E, Step 3, and the security warning). The task creation protocol, gating checks, and 'never skip' warnings add significant overhead. Many instructions are meta-commentary about execution discipline ('If you catch yourself wanting to skip a step, STOP') rather than actionable content.

1 / 3

Actionability

The skill provides fully executable bash commands, specific GraphQL queries, exact gh CLI invocations, concrete commit message formats, and precise decision logic with clear thresholds. Every step has copy-paste ready commands and explicit completion checks.

3 / 3

Workflow Clarity

The multi-step workflow is exceptionally well-sequenced with explicit gate checks between steps, a clear execution order table, validation checkpoints at every stage, feedback loops (Step 3 failing resets to Step 2), and a success streak requirement (5 consecutive clean passes). The decision matrix in 2E is precise with all three signals clearly defined.

3 / 3

Progressive Disclosure

The entire skill is a monolithic wall of text with no references to supporting files. The duplicated GraphQL query blocks, the detailed security warning, and the extensive execution protocol could all be extracted into referenced files. With no bundle files provided, there's no structural decomposition at all despite the content's complexity warranting it.

1 / 3

Total

8

/

12

Passed

Description

60%

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 reasonably specific about what the skill does—orchestrating a review-fix-review loop across sub-skills—but lacks an explicit 'Use when...' clause that would help Claude distinguish when to select this orchestrator skill versus the individual sub-skills. Adding explicit trigger guidance and more natural user-facing keywords would strengthen it.

Suggestions

Add an explicit 'Use when...' clause, e.g., 'Use when the user wants a full automated PR review-and-fix cycle, or asks to iterate on a pull request until all issues are resolved.'

Include natural keyword variations users might say, such as 'pull request', 'iterate', 'review loop', 'clean PR', or 'automated review'.

Clarify the distinction from the individual sub-skills (code-review, address-pr-comments, fix-ci-tests) by emphasizing this is the orchestrator/coordinator that runs them in sequence.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions: self-review a PR, fix all issues, re-review in a loop, and names the coordinated sub-skills (code-review, address-pr-comments, fix-ci-tests).

3 / 3

Completeness

Clearly answers 'what' (self-review a PR, fix issues, re-review in a loop, coordinate sub-skills), but lacks an explicit 'Use when...' clause specifying when Claude should select this skill over others.

2 / 3

Trigger Term Quality

Includes relevant terms like 'PR', 'review', 'fix', 'CI tests', and 'pr-comments', but misses common user variations like 'pull request', 'code review loop', 'iterate on PR', or 'clean up PR'. Users might say 'review my PR and fix issues' but the description doesn't fully cover natural phrasing.

2 / 3

Distinctiveness Conflict Risk

The description mentions coordinating code-review, address-pr-comments, and fix-ci-tests skills, which helps distinguish it as an orchestrator, but it could still overlap with those individual skills since it shares similar trigger terms like 'PR', 'review', and 'fix'.

2 / 3

Total

9

/

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.

Validation10 / 11 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

10

/

11

Passed

Repository
DataDog/rshell
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.