self-improvement

Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.

Quality

79%

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 ./configs/microservice/bff-service/configs/agent-skills/clawhub/self-improvement/SKILL.md

Quality

Content

75%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 executable templates and well-sequenced workflows backed by real bundle files, but it is noticeably verbose and repetitive, with several sections restating the same triggers and guidance. Tightening duplication and trimming over-explained taxonomies would meaningfully improve token efficiency.

Suggestions

Consolidate the repeated trigger/action tables (Quick Reference, Promotion Targets, Detection Triggers, Multi-Agent Support) into a single canonical section to cut redundancy and tokens.

Move the Priority Guidelines, Area Tags, and Best Practices taxonomies into a reference file, keeping only the essential inline summary.

Add an automated validation step (e.g. a script that checks entry ID uniqueness and Status consistency) to upgrade workflow clarity from manual to verified checkpoints.

Dimension	Reasoning	Score
Conciseness	The ~640-line body is noticeably verbose with redundant sections (Quick Reference, Promotion Targets, Detection Triggers, Best Practices, Multi-Agent Support repeat the same triggers/actions multiple times) and some over-explanation of concepts like priority levels and area tags that add length without proportional value.	3 / 5
Actionability	Provides fully executable, copy-paste-ready bash snippets for initialization, ID generation, grep-based review, and hook JSON configs, plus complete markdown templates for every entry type with concrete field-by-field guidance.	5 / 5
Workflow Clarity	Multi-step workflows (logging, promotion, simplify-and-harden ingestion, skill extraction) are clearly sequenced with ID/status state transitions and review checkpoints; however validation checkpoints rely on manual status flips rather than automated verification, leaving a minor gap.	4 / 5
Progressive Disclosure	Good structure with one-level-deep references to real files (references/openclaw-integration.md, references/hooks-setup.md, assets/SKILL-TEMPLATE.md, scripts/*.sh) clearly signaled inline; minor gaps in that substantial inline material (Promotion Targets, Detection Triggers) could itself live in a reference file.	4 / 5
	Total	16 / 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 what the skill does and when to use it via a concrete numbered trigger list with natural user phrases. Minor gaps in action breadth and a slight overlap risk with general planning skills keep it just below perfect.

Suggestions

Add a couple more concrete action verbs (e.g. 'promotes important learnings to project memory') to broaden the specificity of capabilities.

Tighten the final 'review learnings before major tasks' clause to a distinct trigger phrase that reduces overlap with general planning/note-taking skills.

Dimension	Reasoning	Score
Specificity	Names concrete actions ('Captures learnings, errors, and corrections', 'review learnings') and several trigger conditions, but stops short of a comprehensive list of distinct capabilities, leaving minor coverage gaps.	4 / 5
Completeness	Explicitly states both the 'what' ('Captures learnings, errors, and corrections to enable continuous improvement') and the 'when' via a numbered 'Use when' list of six concrete triggers plus a review cadence, satisfying both halves explicitly.	5 / 5
Trigger Term Quality	Embeds natural user phrases like 'No, that's wrong...', 'Actually...', and clear failure signals; good keyword coverage with a few common variations (e.g. synonyms for 'correction') missing.	4 / 5
Distinctiveness Conflict Risk	The self-improvement niche with specific correction/error triggers is mostly distinct, though the broad 'review learnings before major tasks' phrasing could overlap marginally with general note-taking or planning skills.	4 / 5
	Total	17 / 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
skill_md_line_count	SKILL.md is long (645 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: UnicomAI/wanwu
Path: configs/microservice/bff-service/configs/agent-skills/clawhub/self-improvement/SKILL.md
Commit: 99df6e4

Reviewed: 3 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.