Name: self-improvement
Rating: 68 (1 reviews)
Author: pskoett

self-improvement

Captures learnings, errors, corrections, and feature requests to enable continuous improvement. Use when: (1) User corrects Claude ('No, that's wrong...', 'Actually...'), (2) User requests a capability that doesn't exist, (3) Claude realizes its knowledge is outdated or incorrect, (4) A better approach is discovered for a recurring task, (5) Receiving a Handoff block from self-healing (a recurring verified heal with Recurrence-Count at least 3) to distill into a memory file or new skill. For ACTIVE runtime failures where the agent needs to apply and verify a fix mid-task, use `self-healing` instead (it files HEAL- entries with proof; self-improvement promotes accumulated patterns). Also review learnings before major tasks. For CI-only/headless learning capture, use self-improvement-ci.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

The canonical home for this skill is self-improvement in pskoett/pskoett-ai-skills

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.

Highly actionable content with concrete templates, exact formats, and executable commands, plus well-signaled one-level-deep references to real bundle files. Its main weakness is length and redundancy — several sections repeat triggers and guidance that could be consolidated or moved to references.

Suggestions

Consolidate the 'Detection Triggers' section with the Quick Reference table so trigger phrases live in one place, removing the duplicate listing.

Move the per-agent Multi-Agent Support detail (Codex/Copilot/OpenClaw blocks) into references/ and keep only a short pointer table in the body, mirroring how openclaw-integration.md is already handled.

Either reference references/examples.md from the body (e.g., under Logging Format or Skill Extraction) or drop it from the bundle so the file listing matches what is actually navigated to.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and well-structured, but the ~570-line body repeats material across sections (e.g., detection triggers appear in both the Quick Reference table and the 'Detection Triggers' section, and Multi-Agent Support overlaps 'Agent-Agnostic Guidance') that could be tightened.	3 / 5
Actionability	Provides fully copy-paste-ready markdown templates for LEARNINGS/ERRORS/FEATURE_REQUESTS entries, exact ID format ('TYPE-YYYYMMDD-XXX'), concrete grep commands, and an extraction helper with dry-run — concrete and executable throughout.	5 / 5
Workflow Clarity	Multi-step processes (Resolving Entries, Promotion, Simplify & Harden ingestion, Skill Extraction) are clearly numbered with a verification checkpoint in extraction ('Verify: Read skill in fresh session') and search-first dedupe, though validation is lighter for the core append-only logging steps.	4 / 5
Progressive Disclosure	Main body keeps core workflow inline and defers detail to real one-level-deep files (references/openclaw-integration.md, references/hooks-setup.md) and assets/scripts that all exist, though the inlined Multi-Agent and Best Practices sections and the unreferenced references/examples.md leave minor organization gaps.	4 / 5
	Total	16 / 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 clearly states capabilities and provides an explicit numbered 'Use when' trigger list, while actively disambiguating from the related self-healing and CI variants. Only minor room to add a few more natural synonyms or file extensions.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Captures learnings, errors, corrections, and feature requests' plus 'distill into a memory file or new skill' and 'promotes accumulated patterns' — giving comprehensive coverage of what the skill does.	5 / 5
Completeness	Explicitly answers both what ('Captures learnings, errors, corrections, and feature requests to enable continuous improvement') and when (a numbered 'Use when:' list of five concrete triggers plus 'review learnings before major tasks').	5 / 5
Trigger Term Quality	Includes natural quoted trigger phrases ('No, that's wrong...', 'Actually...') and terms like 'outdated', 'better approach', 'recurring task', but lacks file-extension or a few common synonyms; good but not fully comprehensive.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche and explicitly disambiguates from sibling skills — 'For ACTIVE runtime failures... use `self-healing` instead' and 'For CI-only/headless learning capture, use self-improvement-ci' — minimizing wrong-skill triggers.	5 / 5
	Total	19 / 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
skill_md_line_count	SKILL.md is long (589 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 1 suspicious	Warning

	Total	13 / 16 Passed

Repository: pskoett/pskoett-ai-skills
Path: plugin/skills/self-improvement/SKILL.md
Commit: 8a71d70

Reviewed: 12 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.