a-evolve

Apply A-Evolve's agentic evolution methodology to improve AI agent performance across runs. Use when the user wants to diagnose agent failures, generate targeted skills from error patterns, evolve system prompts, or accumulate episodic knowledge. Works standalone or inside AutoResearchClaw pipelines. Triggers on: "evolve", "self-improve", "diagnose failures", "generate skills from errors", "what went wrong and how to fix it", or any mention of A-Evolve.

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

92%Weight 40%Scale 1-3

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, concise, and presents a clear validated workflow. Its one weakness is progressive disclosure: everything lives in one ~190-line SKILL.md with no reference files, so peripheral material is inlined rather than split out for navigation.

Suggestions

Move the AutoResearchClaw stage-mapping table and the MetaClaw relationship section into a separate reference file (e.g. references/autoresearchclaw-integration.md) and link to it from the body, keeping the core loop front and center.

Extract the full evolved-skill example and knowledge-entry JSON schema into a reference (e.g. references/mutation-templates.md) so the main SKILL.md stays a lean overview.

Trim the inlined example SKILL.md and observation block to shorter illustrations, pointing to a templates reference for the full versions.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence — it never explains basic concepts (logs, errors, pagination) and every section earns its tokens, matching the score-3 anchor for lean, efficient content.	3 / 3
Actionability	Provides executable, copy-paste-ready artifacts: a concrete observation format, a full SKILL.md example, a prompt-patch template, a knowledge-entry JSON schema, and a file-placement table with explicit paths.	3 / 3
Workflow Clarity	The Solve→Observe→Evolve→Gate→Reload loop is clearly sequenced, and the Gate step supplies an explicit validation checklist (Specificity, Testability, Blast radius, Consistency) with a refine-or-discard feedback loop, matching the score-3 anchor.	3 / 3
Progressive Disclosure	The skill is a single ~190-line file with no bundle references; while well-organized into sections, material such as the AutoResearchClaw stage mapping and MetaClaw relationship is inlined that could appropriately live in separate reference files, matching the score-2 anchor for content that should be separate being inline.	2 / 3
	Total	11 / 12 Passed

Description

100%Weight 40%Scale 1-3

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 across all dimensions: it names concrete capabilities, provides explicit natural-language triggers, covers both what and when, and carves out a distinct niche. No missing trigger guidance or vague fluff.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'diagnose agent failures', 'generate targeted skills from error patterns', 'evolve system prompts', 'accumulate episodic knowledge' — matching the score-3 anchor for multiple specific concrete actions.	3 / 3
Completeness	Explicitly answers both what (improve agent performance via A-Evolve methodology) and when (an explicit 'Use when the user wants to...' clause plus 'Triggers on:'), matching the score-3 anchor.	3 / 3
Trigger Term Quality	Provides natural trigger phrases users would actually say ('evolve', 'self-improve', 'diagnose failures', 'what went wrong and how to fix it') plus 'any mention of A-Evolve', giving good coverage rather than technical jargon.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche tied to A-Evolve and AutoResearchClaw with distinct triggers, making it unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: aiming-lab/AutoResearchClaw
Path: .claude/skills/a-evolve/SKILL.md
Commit: e2e23c9

Reviewed: 6 days 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.