self-improving-agent

Curate Claude Code's auto-memory into durable project knowledge. Analyze MEMORY.md for patterns, promote proven learnings to CLAUDE.md and .claude/rules/, extract recurring solutions into reusable skills. Use when: (1) reviewing what Claude has learned about your project, (2) graduating a pattern from notes to enforced rules, (3) turning a debugging solution into a skill, (4) checking memory health and capacity.

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

—

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./engineering-team/self-improving-agent/skills/self-improving-agent/SKILL.md

Quality

Content

50%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.

A well-structured README-style plugin overview that reads more as marketing/documentation than operational skill instruction. It explains concepts and lists commands but provides little executable guidance for the actual curation workflow and no progressive disclosure via bundle files.

Suggestions

Replace conceptual prose (e.g. 'Auto-memory is capture, not curation', 'Promotion = graduation') with lean operational instructions; let the tables and commands carry the weight.

Add concrete, executable steps and validation checkpoints for destructive operations — e.g. how to run a memory-review, how to confirm a promotion candidate before removing its MEMORY.md entry, and how to verify a promoted rule loads.

Move agent, hook, and platform-support detail into reference files under references/ and link to them from the body so SKILL.md stays a concise overview.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with clear tables, but the ASCII memory-stack diagram and conceptual prose ('Auto-memory is capture, not curation', 'Promotion = graduation') pad the body with explanation a competent Claude largely doesn't need.	3 / 5
Actionability	Install commands and a YAML rules snippet are concrete and executable, but the core curation tasks (how to analyze MEMORY.md, how to actually promote an entry) are described at a high level without executable steps or commands.	3 / 5
Workflow Clarity	A promotion lifecycle (discover → flag → approve → promote → remove) is sequenced, but it is conceptual rather than operational and lacks validation checkpoints; destructive removal of MEMORY.md entries triggers the workflow_clarity cap at 3.	3 / 5
Progressive Disclosure	Sections are clearly headed and the body is navigable, but there are no bundle files (references/scripts/assets) and inline content such as agent, hook, and platform-support detail that could be split out is all in one ~160-line file.	3 / 5
	Total	12 / 20 Passed

Description

92%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.

A strong, third-person description that pairs concrete capabilities with explicit, multi-scenario trigger guidance. Minor room to add a few more colloquial trigger synonyms.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Analyze MEMORY.md for patterns', 'promote proven learnings to CLAUDE.md and .claude/rules/', 'extract recurring solutions into reusable skills' — giving comprehensive coverage of capabilities.	5 / 5
Completeness	Explicitly answers both 'what' (curate, analyze, promote, extract) and 'when' with four concrete numbered trigger scenarios, matching the top anchor.	5 / 5
Trigger Term Quality	The 'Use when' clause offers four natural scenario triggers ('reviewing what Claude has learned', 'graduating a pattern from notes to enforced rules', 'turning a debugging solution into a skill', 'checking memory health and capacity'), but a few common phrasings/synonyms (e.g. 'clean up memory', 'remember this') are absent.	4 / 5
Distinctiveness Conflict Risk	Clear niche — Claude Code auto-memory curation — anchored to specific artifacts (MEMORY.md, CLAUDE.md, .claude/rules/), making it highly distinguishable with minimal conflict risk.	5 / 5
	Total	19 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: alirezarezvani/claude-skills
Path: engineering-team/self-improving-agent/skills/self-improving-agent/SKILL.md
Commit: 19392f7

Reviewed: ~9 hours 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.