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wrong...', 'Actually...'), (2) User requests a capability that doesn't exist, (3) Claude realizes its knowledge is outdated or incorrect, (4) A better approach is discovered for a recurring task, (5) Receiving a Handoff block from self-healing (a recurring verified heal at Recurrence-Count >= 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.

2.48x

Quality

75%

Does it follow best practices?

Impact

97%

2.48x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

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

Evaluation results

100%

Documenting Development Incidents and Requests

Learning entry format and file routing

Criteria

Without context

With context

.learnings dir created

100%

Correction in LEARNINGS.md

100%

LRN ID format

100%

Correction category

100%

Learning required fields

100%

Learning sections present

100%

Valid area tag

100%

Status is pending

100%

Feature request in FEATURE_REQUESTS.md

100%

FEAT ID format

100%

Feature request required fields

100%

Feature request sections

100%

Complexity estimate value

100%

Source metadata field

100%

93%

24%

Consolidating Project Learnings into Permanent Documentation

Learning promotion to correct project files

Criteria

Without context

With context

API regen → AGENTS.md

100%

pnpm → CLAUDE.md

100%

AGENTS.md content is concise

100%

CLAUDE.md content is concise

100%

Status updated to promoted

100%

Promoted field added

60%

100%

Both promotable entries updated

100%

No README.md as target

100%

No SOUL.md or TOOLS.md

100%

Unpromotable entries unchanged

Promotion report present

100%

Destination reasoning

100%

50%

Packaging a Proven Pattern as a Reusable Skill

Skill extraction naming and structure

Criteria

Without context

With context

skills/ folder created

100%

Skill name is lowercase-hyphens

100%

SKILL.md present (not README.md)

100%

YAML frontmatter present

100%

Frontmatter name matches folder

100%

Description mentions trigger

100%

Status updated to promoted_to_skill

100%

Skill-Path added to learning

28%

100%

Other entry unchanged

100%

Skill log present

100%

Repository: pskoett/pskoett-ai-skills
Commit: f6c5d7b

Evaluated: 11 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Documenting Development Incidents and Requests Consolidating Project Learnings into Permanent Documentation Packaging a Proven Pattern 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.