auto-review-loop
Autonomous multi-round research review loop. Repeatedly reviews using Claude Code via claude-review MCP, implements fixes, and re-reviews until positive assessment or max rounds reached. Use when user says "auto review loop", "review until it passes", or wants autonomous iterative improvement.
94
92%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong skill description that clearly communicates a specific autonomous review workflow with well-defined actions, termination conditions, and explicit trigger phrases. It uses third person voice correctly and provides enough detail to distinguish it from simpler review-related skills. The only minor weakness is that 'research review' could be slightly ambiguous (academic research vs code review), but the context of 'Claude Code' and 'implements fixes' clarifies the intent.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'reviews using Claude Code via claude-review MCP', 'implements fixes', 're-reviews until positive assessment or max rounds reached'. Describes a clear multi-step process with termination conditions. | 3 / 3 |
Completeness | Clearly answers both 'what' (autonomous multi-round research review loop that reviews, fixes, and re-reviews) and 'when' (explicit 'Use when' clause with specific trigger phrases like 'auto review loop' and 'review until it passes'). | 3 / 3 |
Trigger Term Quality | Includes natural trigger phrases users would say: 'auto review loop', 'review until it passes', 'autonomous iterative improvement'. These are realistic phrases a user would use when requesting this specific workflow. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: autonomous iterative review loop using a specific MCP tool (claude-review). The combination of 'autonomous', 'multi-round', 'review loop', and the specific MCP reference makes it unlikely to conflict with simple code review or one-shot review skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a high-quality, complex skill that provides a thorough autonomous review loop workflow with excellent actionability and workflow clarity. The multi-phase structure with explicit state persistence, crash recovery, and human checkpoint options demonstrates sophisticated design. The main weakness is moderate verbosity — some instructions are repeated (polling pattern appears twice verbatim) and a few sections could be tightened without losing clarity.
Suggestions
Deduplicate the polling instructions (save jobId → poll review_status → extract response) by defining it once as a named pattern and referencing it in Phase A and the Round 2+ template, rather than repeating the full description verbatim.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly long (~200+ lines) and includes some redundant explanations (e.g., repeating the polling instructions verbatim for round 2+, explaining fallback paths multiple times). However, most content is necessary given the complexity of the multi-round autonomous loop. Some tightening is possible but it's not egregiously verbose. | 2 / 3 |
Actionability | The skill provides concrete MCP tool calls with specific prompt templates, exact JSON schemas for state persistence, specific file paths, clear parsing instructions for scores/verdicts, and copy-paste ready markdown templates for documentation. The human checkpoint interaction patterns are specific with exact trigger words. | 3 / 3 |
Workflow Clarity | The workflow is exceptionally well-structured with clear phases (A through E), explicit stop conditions, validation checkpoints (score thresholds, verdict parsing), state persistence for crash recovery, and feedback loops (review → fix → re-review). The initialization section handles multiple edge cases (fresh start, resume, stale state) with clear decision logic. | 3 / 3 |
Progressive Disclosure | The skill is well-organized with clear sections (Constants, State Persistence, Workflow phases, Key Rules, Prompt Templates). It references shared protocols via links (output versioning, manifest, language) rather than inlining them. The collapsible details block for raw reviewer responses is a nice touch. Content is appropriately structured for its complexity. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Commit
700fbe2
- 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.