Self improvement
Multi-agent team for Claude Code: 7 AI specialists (strategy, dev, security, sales, design, data, research) that collaborate and challenge you. Dev-cycle workflow with machine-verifiable completion.
npx -y skills add ToruAI/toru-claude-agents --skill self-improvementAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 15 stars15 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Turn friction you just hit into a concrete change to the workflow, so the same problem doesn't recur.
SKILL.md
2.9 KB, as published. Nobody here has run it
Self-Improvement Protocol
You are responsible for making yourself and the system better over time.
The Improvement Loop
1. NOTICE - Identify friction, repetition, or failure
2. ANALYZE - Why did this happen? What's the root cause?
3. DESIGN - What's the minimal fix?
4. IMPLEMENT - Create skill/hook/agent update
5. TEST - Verify it works
6. DOCUMENT - Log in megg
7. SHARE - Tell the user if significant
What to Improve
Process Friction
- Repeated manual steps → automate with hooks
- Same context loading → create skill
- Unclear delegation → improve agent descriptions
- Lost decisions → strengthen megg discipline
Quality Issues
- Bugs that reach review → add validation hooks
- Incomplete work → improve checklists in skills
- Scope creep → add explicit boundaries
- Vision drift → strengthen context protocols
Efficiency
- Slow operations → use subagents for parallel work
- Context bloat → progressive disclosure in skills
- Repeated research → cache in megg or skill
When to Improve
After every significant task:
- What went well?
- What was harder than expected?
- What would I do differently?
When you notice:
- Same thing done 3+ times → automate
- Decision made twice → document
- Error repeated → add guard
Periodically:
- Review megg decisions - patterns?
- Check skills - still accurate?
- Audit hooks - still needed?
Where to Put Improvements
| Scope | Location | When |
|---|---|---|
| Personal | ~/.claude/skills/ | Your workflows only |
| Project | .claude/skills/ | Team needs it |
| System | Update agent definitions | Core behavior change |
Documentation Pattern
When creating improvement, log it:
## YYYY-MM-DD - Improvement: [Name]
**Friction:** What problem was encountered
**Root Cause:** Why it happened
**Solution:** What was created/changed
**Location:** Where the fix lives
**Verification:** How to know it works
Improvement Ideas Backlog
When you notice something but can't fix now:
- Add to
~/.claude/skills/self-improvement/ideas.md - Include: friction, proposed fix, priority
- Review periodically
Anti-Patterns
- Improving without logging - Others (including future you) won't know
- Premature optimization - Fix real problems, not theoretical ones
- Over-automation - Some things should stay manual
- Isolated improvements - Share learnings through megg
Philosophy
"Every friction point is an improvement opportunity."
Good systems get better over time. You're not just executing tasks - you're building the machine that executes tasks.
Questions to ask:
- "How would I want this to work next time?"
- "What would make the user unnecessary here?"
- "What would the ideal workflow look like?"
Then build toward that.