auto-review-loop-llm

Autonomous research review loop using any OpenAI-compatible LLM API. Configure via llm-chat MCP server or environment variables. Trigger with "auto review loop llm" or "llm review".

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 ./skills/auto-review-loop-llm/SKILL.md

Quality

Content

63%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 highly actionable with concrete API call patterns, state schemas, and a clearly sequenced review loop with feedback checkpoints. Its weaknesses are duplicated curl/prompt blocks that hurt token efficiency, an executable bug in the curl JSON quoting, and monolithic inlining with references to missing shared-references files.

Suggestions

Remove the duplicated curl fallback block (it appears both in "API Call Method" and in Phase A) and reference one canonical version to improve conciseness.

Fix the curl JSON so environment variables expand — either build the JSON body with a here-doc / double-quoted string or substitute variables before sending, since ${LLM_MODEL} inside single quotes will not interpolate.

Split the provider table and the round 2+ prompt template into reference files (e.g., references/providers.md, references/prompts.md) and either ship the referenced ../shared-references/*.md files or inline that guidance so no links point to absent files.

Dimension	Reasoning	Score
Conciseness	The body is mostly concrete config, commands, and prompt templates without explaining concepts Claude already knows, but the curl fallback is duplicated verbatim (general "API Call Method" and again in Phase A) and the prompt structure repeats, so it could be tightened.	3 / 5
Actionability	Provides copy-paste-ready MCP tool invocations, curl commands, JSON state schema, and prompt templates with an exact STOP condition, but the curl JSON uses ${VAR} inside single quotes (which will not expand) and prompts contain unfilled placeholders, leaving minor executable gaps.	4 / 5
Workflow Clarity	Sequences Initialization through Phases A-E to Termination with an explicit STOP check, per-round state persistence, and a built-in review->fix->re-review feedback loop, satisfying the feedback-loop requirement for batch operations, though pre-re-review validation of fixes is only implicit.	4 / 5
Progressive Disclosure	No bundle files exist, yet the body references ../shared-references/*.md (external-cadence, output-versioning, output-manifest, output-language) that are not present, and the ~260-line file inlines provider tables, JSON schemas, full prompts, and curl blocks that could be split into separate reference files.	3 / 5
	Total	14 / 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 strong: it explicitly answers both what the skill does and when to use it with concrete trigger phrases, and names specific configuration mechanisms. The main weakness is slightly narrow trigger coverage and a single-capability framing that limits breadth of distinct actions.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete mechanisms ("any OpenAI-compatible LLM API", "llm-chat MCP server or environment variables", explicit trigger phrases), but describes one loop capability rather than enumerating multiple distinct actions, leaving minor coverage gaps.	4 / 5
Completeness	Clearly states both what it does ("Autonomous research review loop using any OpenAI-compatible LLM API") and when to trigger it ("Trigger with \"auto review loop llm\" or \"llm review\""), matching the anchor that requires both with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes explicit natural trigger phrases ("Trigger with \"auto review loop llm\" or \"llm review\"") that a user might actually say, but coverage is narrow with common synonyms like "review my research" missing.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (autonomous LLM-driven research review loop) with specific triggers, but "llm review" is generic enough to risk minor overlap with general review skills.	4 / 5
	Total	17 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 4 suspicious	Warning

	Total	13 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/auto-review-loop-llm/SKILL.md
Commit: f4f20f9

Reviewed: 1 day 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.