CtrlK
BlogDocsLog inGet started
Tessl Logo

auto-review-loop

Autonomous multi-round research review loop. Repeatedly reviews using Gemini via gemini-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.

68

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Critical

Do not install without reviewing

SKILL.md
Quality
Evals
Security

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 body is highly actionable and workflow-clear with concrete MCP calls and a robust state-recovery loop, scoring top marks on those dimensions. Conciseness is the weakest point due to a verbatim-duplicated SCOPE LIMITS block across two prompt templates.

Suggestions

De-duplicate the SCOPE LIMITS block: define it once and reference it from the Round 2+ template to reclaim tokens without losing the guidance.

Consider moving the large Round 1 and Round 2+ prompt templates into a references/ file so SKILL.md stays a lean overview pointing to the verbatim prompts.

Tighten the Constants section by removing parenthetical legacy-fallback notes that repeat the same path-twice pattern already shown in the Workflow steps.

DimensionReasoningScore

Conciseness

Mostly efficient operational guidance, but the SCOPE LIMITS block is duplicated verbatim across the Round 1 and Round 2+ prompt templates and several constants are re-explained, adding tokens a competent agent could infer or deduplicate.

3 / 5

Actionability

Provides concrete, executable MCP tool calls with parameters, bounded polling logic, an exact state JSON schema, and copy-ready markdown templates covering the common cases.

5 / 5

Workflow Clarity

Phases A–E are clearly sequenced with an explicit STOP CONDITION, state-persistence/recovery checkpoints, and error-handling guidance (stale-state purge, large-file fallback), giving a complete feedback loop for this batch/iterative operation.

5 / 5

Progressive Disclosure

Single-file skill with well-organized sections and one-level-deep references to shared protocols (output-versioning, output-manifest, output-language); no bundle files exist, so structure is appropriate with minor room to externalize the duplicated prompt templates.

4 / 5

Total

17

/

20

Passed

Description

83%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 is specific, third-person, and answers both what the skill does and when to invoke it with concrete trigger phrases. Trigger-term and distinctiveness are strong but fall just short of fully comprehensive.

DimensionReasoningScore

Specificity

Names the domain (autonomous research review loop) and several concrete actions ('reviews using Gemini via gemini-review MCP', 'implements fixes', 're-reviews'), with minor coverage gaps around what kinds of fixes get implemented.

4 / 5

Completeness

Clearly states what it does ('Repeatedly reviews... implements fixes, and re-reviews until positive assessment or max rounds reached') and an explicit 'Use when...' clause with concrete trigger phrases, answering both what and when.

5 / 5

Trigger Term Quality

Includes natural phrases users would say ('auto review loop', 'review until it passes', 'autonomous iterative improvement'); a few common synonyms or shorthand variations are missing, keeping it just below comprehensive.

4 / 5

Distinctiveness Conflict Risk

Has a clear niche (Gemini-as-reviewer autonomous loop) with distinct triggers; minor overlap risk with generic review or skills-codex review skills, so not a 5.

4 / 5

Total

17

/

20

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 3 suspicious

Warning

Total

15

/

16

Passed

Repository
wanshuiyin/Auto-claude-code-research-in-sleep
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.