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User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task.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. For CI-only/headless learning capture, use self-improvement-ci.

1.86x

Quality

72%

Does it follow best practices?

Impact

93%

1.86x

Average score across 6 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/self-improvement/SKILL.md

Evaluation results

97%

78%

Documenting Developer Incidents and Gaps

Learning entry logging format

Criteria

Without context

With context

Error ID format

100%

Error has code block

100%

Error context section

50%

100%

Error reproducible field

100%

Error priority high

100%

Learning ID format

100%

Learning category correction

25%

100%

Learning ISO-8601 timestamp

33%

100%

Learning source field

100%

Learning Suggested Action

60%

100%

Feature request ID format

100%

Feature request priority medium

100%

Feature request complexity

100%

Feature request frequency

100%

All status pending

40%

Area tags used

25%

100%

89%

Elevating Key Learnings to Project Memory

Promoting learnings to project memory

Criteria

Without context

With context

LRN-001 status promoted

100%

LRN-001 Promoted field

100%

LRN-002 status promoted

100%

LRN-002 Promoted field

100%

Payments rule in CLAUDE.md

100%

Test env rule in AGENTS.md

100%

CLAUDE.md is concise

50%

AGENTS.md is actionable

62%

87%

Correct promotion target for payments

100%

Correct promotion target for tests

100%

Original entries preserved

100%

No verbose incident write-up in promoted files

25%

95%

45%

Packaging a Recurring Fix as a Reusable Skill

Skill extraction from learning entry

Criteria

Without context

With context

Skill name lowercase-hyphen

50%

100%

Name matches folder

100%

YAML frontmatter present

100%

Description is meaningful

100%

No README.md in skill folder

100%

Quick Reference section

100%

Source learning ID recorded

100%

Learning status promoted_to_skill

50%

Skill-Path field added

40%

100%

Skill is self-contained

100%

Skill has solution content

100%

Skill folder structure only has SKILL.md

100%

Repository: pskoett/pskoett-ai-skills
Commit: d6c68fa

Evaluated: 8 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Documenting Developer Incidents and Gaps Elevating Key Learnings to Project Memory Packaging a Recurring Fix as a Reusable Skill

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.