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User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task.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.
81
74%
Does it follow best practices?
Impact
97%
4.21xAverage score across 3 eval scenarios
Passed
No findings from the security scan
Fix and improve this skill with Tessl
tessl review fix ./clawdbot/self-improving-agent/SKILL.mdThe canonical home for this skill is self-improving-agent in jdrhyne/agent-skills
Evaluation results
100%
87%Backend Service Debugging Session
Error entry format and routing
Learnings dir created
0%
100%
Errors routed to ERRORS.md
0%
100%
ERR- prefix in ID
0%
100%
Date in ID
0%
100%
3-char suffix in ID
0%
100%
Priority: high
0%
100%
Status: pending
0%
100%
ISO-8601 Logged timestamp
0%
100%
Error section with content
50%
100%
Context section
50%
100%
Suggested Fix section
50%
100%
Reproducible field
0%
100%
92%
35%Consolidating Developer Learnings into Project Documentation
Promoting learnings to project memory
CLAUDE.md created or updated
100%
100%
AGENTS.md created or updated
0%
100%
Status updated to promoted
100%
100%
Selective promotion
0%
0%
Promoted field added
30%
100%
CLAUDE.md is concise
87%
100%
AGENTS.md has workflow rule
0%
100%
CI workflow rule in AGENTS.md
0%
100%
Correct target for package manager
100%
100%
Project-specific entry not over-promoted
100%
100%
Learnings file preserved
100%
100%
100%
100%Migrating a React Codebase to Modern Patterns
Learning entry format and categories
Learnings routed to LEARNINGS.md
0%
100%
LRN- prefix in ID
0%
100%
Date in ID
0%
100%
3-char suffix in ID
0%
100%
Category in heading
0%
100%
ISO-8601 Logged timestamp
0%
100%
Valid Priority value
0%
100%
Status: pending
0%
100%
Valid Area tag
0%
100%
Summary section
0%
100%
Details section
0%
100%
Suggested Action section
0%
100%
- Repository
- bap-jorkim/agent-skills-fork-feb-25
- Commit
f4ed46f
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
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.