Self-Improvement Skill

Log learnings and errors to markdown files for continuous improvement. Coding agents can later process these into fixes, and important learnings get promoted to project memory.

For installation commands, see references/installation.md. For optional agent activation and hooks, see references/hooks-setup.md. Read these only when installing or configuring the skill.

Pair with self-healing: self-healing is the active runtime recovery primitive — it diagnoses, patches, verifies, and files HEAL- entries to .learnings/HEALS.md when something breaks mid-task. Self-improvement (this skill) is the passive accumulation and promotion layer — it logs corrections, knowledge gaps, and feature requests, and promotes recurring heal handoffs to permanent memory. They share .learnings/ but write to different files; verify discipline lives in self-healing, promotion logic lives here.

Quick Reference

Setup

Create .learnings/ directory in project root if it doesn't exist:

mkdir -p .learnings

Copy the file templates from assets/ (LEARNINGS.md, ERRORS.md, FEATURE_REQUESTS.md) or create files with headers.

Logging Format

Learning Entry

Append to .learnings/LEARNINGS.md:

Every saved learning gets maintenance fields at capture, even if never promoted. Use references/context-maintenance.md; keep unknowns explicit.

## [LRN-YYYYMMDD-XXX] category

**Logged**: ISO-8601 timestamp
**Priority**: low | medium | high | critical
**Status**: pending
**Area**: frontend | backend | infra | tests | docs | config

### Summary
One-line description of what was learned

### Details
Full context: what happened, what was wrong, what's correct

### Suggested Action
Specific fix or improvement to make

### Metadata
- Source: conversation | error | user_feedback
- Related Files: path/to/file.ext
- Tags: tag1, tag2
- See Also: LRN-20250110-001 (if related to existing entry)
- Pattern-Key: simplify.dead_code | harden.input_validation (optional, for recurring-pattern tracking)
- Recurrence-Count: 1 (optional)
- First-Seen: 2025-01-15 (optional)
- Last-Seen: 2025-01-15 (optional)

### Context Maintenance
- Decay-Type: reality | decision | dependency | relevance | unknown
- Decay-Rate: fast | medium | slow | durable | unknown (brief basis)
- Authority: source or unknown; Owner: known role or unknown
- Revalidation: method; invalidating triggers; when to check; unavailable-source behavior
- Validation: pending | verified | unresolved; Last-Validated: date or none; Evidence: checked source or none
- Disposition: unassessed | retain | revise | externalize | retire; Application: proposed | applied

---

Error Entry

Append to .learnings/ERRORS.md:

## [ERR-YYYYMMDD-XXX] skill_or_command_name

**Logged**: ISO-8601 timestamp
**Priority**: high
**Status**: pending
**Area**: frontend | backend | infra | tests | docs | config

### Summary
Brief description of what failed

### Error

Actual error message or output

### Context
- Command/operation attempted
- Input or parameters used
- Environment details if relevant

### Suggested Fix
If identifiable, what might resolve this

### Metadata
- Reproducible: yes | no | unknown
- Related Files: path/to/file.ext
- See Also: ERR-20250110-001 (if recurring)

---

Feature Request Entry

Append to .learnings/FEATURE_REQUESTS.md:

## [FEAT-YYYYMMDD-XXX] capability_name

**Logged**: ISO-8601 timestamp
**Priority**: medium
**Status**: pending
**Area**: frontend | backend | infra | tests | docs | config

### Requested Capability
What the user wanted to do

### User Context
Why they needed it, what problem they're solving

### Complexity Estimate
simple | medium | complex

### Suggested Implementation
How this could be built, what it might extend

### Metadata
- Frequency: first_time | recurring
- Related Features: existing_feature_name

---

ID Generation

Format: TYPE-YYYYMMDD-XXX

• TYPE: LRN (learning), ERR (error), FEAT (feature)
• YYYYMMDD: Current date
• XXX: Sequential number or random 3 chars (e.g., 001, A7B)

Examples: LRN-20250115-001, ERR-20250115-A3F, FEAT-20250115-002

Resolving Entries

When an issue is fixed, update the entry:

1. Change **Status**: pending → **Status**: resolved
2. Add resolution block after Metadata:

### Resolution
- **Resolved**: 2025-01-16T09:00:00Z
- **Commit/PR**: abc123 or #42
- **Notes**: Brief description of what was done

Other status values:

• in_progress - Actively being worked on
• wont_fix - Decided not to address (add reason in Resolution notes)
• promoted - Elevated to CLAUDE.md, AGENTS.md, or .github/copilot-instructions.md
• promoted_to_skill - Extracted as a reusable skill (see Automatic Skill Extraction)

Promoting to Project Memory

When a learning is broadly applicable (not a one-off fix), promote it to permanent project memory.

When to Promote

• Learning applies across multiple files/features
• Knowledge any contributor (human or AI) should know
• Prevents recurring mistakes
• Documents project-specific conventions

Promotion Targets

How to Promote

The main agent performs this step using the learning entry or the learning-aggregator gap report. No dedicated agent preset is required.

1. Check eligibility and relevance. Apply the Promotion Rule below, read the supporting entries and related files, and check existing instructions. If the rule is already covered, record the existing target rather than duplicating it. Flag stale evidence or conflicting rules for human review; do not silently dismiss entries or overwrite existing guidance.
2. Distill the pattern into one to three actionable, contextual sentences. Choose the appropriate target and section, preserving its organization and any required synchronization across instruction files. Promotion changes instructions, not application code.
3. Show the proposed diff for human approval before writing instructions. Apply only the approved changes. In Copilot, present the diff and ask in chat. Skill edits require explicit scoped approval as described below.
4. Update the source entry after promotion: set **Status**: promoted, record **Promoted**: <target files> and Promoted-Date: YYYY-MM-DD, and retain the source IDs and Pattern-Key. Do not mark an unapplied proposal as promoted. Promotion approval does not authorize committing or publishing.
5. Flag eval candidates for eval-creator when the rule has a clear pass/fail condition. Include the Pattern-Key, rule, target, source IDs, recurrence evidence, assertion, and verification method.

When a gap report supplies a tracker, preserve it as the Pattern-Key in the handoff. Projects using inline provenance can also place <!-- tracker:[pattern-key] source:[entry-IDs] promoted:YYYY-MM-DD eval:[eval-ID] --> above the rule; omit unknown fields. Tracker comments are optional.

Context Maintenance at Promotion

Refine the capture-time contract at promotion; use context-decay when available. Keep metadata in the learning and only the rule or retrieval pointer in instructions. Review does not change recurrence or eligibility; see references/context-maintenance.md. For stale skill guidance, surface a concrete proposal even on the first learning. Skill edits/retirement need explicit scoped approval, even without context-decay; see the same reference.

Promotion Examples

Learning (verbose):

Project uses pnpm workspaces. Attempted npm install but failed. Lock file is pnpm-lock.yaml. Must use pnpm install.

In CLAUDE.md (concise):

## Build & Dependencies
- Package manager: pnpm (not npm) - use `pnpm install`

Learning (verbose):

When modifying API endpoints, must regenerate TypeScript client. Forgetting this causes type mismatches at runtime.

In AGENTS.md (actionable):

## After API Changes
1. Regenerate client: `pnpm run generate:api`
2. Check for type errors: `pnpm tsc --noEmit`

Recurring Pattern Detection

If logging something similar to an existing entry:

1. Search first: grep -r "keyword" .learnings/
2. Link entries: Add **See Also**: ERR-20250110-001 in Metadata
3. Bump priority if issue keeps recurring
4. Consider systemic fix: Recurring issues often indicate:
   • Missing documentation (→ promote to CLAUDE.md or .github/copilot-instructions.md)
   • Missing automation (→ add to AGENTS.md)
   • Architectural problem (→ create tech debt ticket)

Simplify & Harden Feed

Use this workflow to ingest recurring patterns from the simplify-and-harden skill and turn them into durable prompt guidance.

Ingestion Workflow

1. Read simplify_and_harden.learning_loop.candidates from the task summary.
2. For each candidate, use pattern_key as the stable dedupe key.
3. Search .learnings/LEARNINGS.md for an existing entry with that key:
   • grep -n "Pattern-Key: <pattern_key>" .learnings/LEARNINGS.md
4. If found:
   • Increment Recurrence-Count
   • Update Last-Seen
   • Add See Also links to related entries/tasks
5. If not found:
   • Create a new LRN-... entry
   • Set Source: simplify-and-harden
   • Set Pattern-Key, Recurrence-Count: 1, and First-Seen/Last-Seen

Promotion Rule (System Prompt Feedback)

Promote recurring patterns into agent context/system prompt files when all are true:

• Recurrence-Count >= 3
• Seen across at least 2 distinct tasks
• Occurred within a 30-day window

Promotion targets:

• CLAUDE.md
• AGENTS.md
• .github/copilot-instructions.md

This three-condition rule is the single promotion threshold for this skill. The Quick Reference row for self-healing Handoff blocks and the aggregator skills (learning-aggregator, learning-aggregator-ci) all use this same rule.

Write promoted rules as short prevention rules (what to do before/while coding), not long incident write-ups.

Periodic Review

Review .learnings/ at natural breakpoints:

When to Review

• Before starting a new major task
• After completing a feature
• When working in an area with past learnings
• Weekly during active development

Quick Status Check

# Count pending items
grep -h "Status\*\*: pending" .learnings/*.md | wc -l

# List pending high-priority items
grep -B5 "Priority\*\*: high" .learnings/*.md | grep "^## \["

# Find learnings for a specific area
grep -l "Area\*\*: backend" .learnings/*.md

Review Actions

• Resolve fixed items
• Promote applicable learnings
• Link related entries
• Escalate recurring issues
• Review saved learnings, including one-off and unpromoted entries, when their source, decision, dependency, or review point changes. Update the existing contract using references/context-maintenance.md; use context-decay when available. A review is not recurrence or evidence for age-based removal.

Detection Triggers

Automatically log when you notice:

Corrections (→ learning with correction category):

• "No, that's not right..."
• "Actually, it should be..."
• "You're wrong about..."
• "That's outdated..."

Feature Requests (→ feature request):

• "Can you also..."
• "I wish you could..."
• "Is there a way to..."
• "Why can't you..."

Knowledge Gaps (→ learning with knowledge_gap category):

• User provides information you didn't know
• Documentation you referenced is outdated
• API behavior differs from your understanding

Errors (→ error entry):

• Command returns non-zero exit code
• Exception or stack trace
• Unexpected output or behavior
• Timeout or connection failure

Priority Guidelines

Area Tags

Use to filter learnings by codebase region:

Best Practices

1. Log immediately - context is freshest right after the issue
2. Be specific - future agents need to understand quickly
3. Include reproduction steps - especially for errors
4. Link related files - makes fixes easier
5. Suggest concrete fixes - not just "investigate"
6. Use consistent categories - enables filtering
7. Promote selectively - apply the promotion rule; prefer a source-retrieval pointer when a copied fact would quickly become stale
8. Review regularly - stale learnings lose value

Gitignore Options

Keep learnings local (per-developer):

.learnings/

Track learnings in repo (team-wide): Don't add to .gitignore - learnings become shared knowledge.

Hybrid (track templates, ignore entries):

.learnings/*.md
!.learnings/.gitkeep

Hook Integration

Automatic reminders are opt-in. Claude Code, Codex CLI, and Copilot have different event and output contracts; see references/hooks-setup.md. Amp loads skills on demand and uses a separate native plugin for per-turn reminders; see references/amp-setup.md. Reminders do not write learnings or guarantee capture.

Automatic Skill Extraction

When a learning is valuable enough to become a reusable skill, extract it using the provided helper.

Skill Extraction Criteria

A learning qualifies for skill extraction when ANY of these apply:

Extraction Workflow

1. Identify candidate: Learning meets extraction criteria
2. Run helper (or create manually):

   ./skills/self-improvement/scripts/extract-skill.sh skill-name --dry-run
   ./skills/self-improvement/scripts/extract-skill.sh skill-name
   

   Copy
3. Customize SKILL.md: Fill in template with learning content
4. Update learning: Set status to promoted_to_skill, add Skill-Path
5. Verify: Read skill in fresh session to ensure it's self-contained

Manual Extraction

If you prefer manual creation:

1. Create skills/<skill-name>/SKILL.md
2. Use template from assets/SKILL-TEMPLATE.md
3. Follow Agent Skills spec:
   • YAML frontmatter with name and description
   • Name must match folder name
   • No README.md inside skill folder

Extraction Detection Triggers

Watch for these signals that a learning should become a skill:

In conversation:

• "Save this as a skill"
• "I keep running into this"
• "This would be useful for other projects"
• "Remember this pattern"

In learning entries:

• Multiple See Also links (recurring issue)
• High priority + resolved status
• Category: best_practice with broad applicability
• User feedback praising the solution

Skill Quality Gates

Before extraction, verify:

• Solution is tested and working
• Description is clear without original context
• Code examples are self-contained
• No project-specific hardcoded values
• Follows skill naming conventions (lowercase, hyphens)

Multi-Agent Support

Regardless of agent, apply self-improvement when you:

1. Discover something non-obvious - solution wasn't immediate
2. Correct yourself - initial approach was wrong
3. Learn project conventions - discovered undocumented patterns
4. Hit unexpected errors - especially if diagnosis was difficult
5. Find better approaches - improved on your original solution

Copilot Chat Integration

For Copilot users, add this to your prompts when relevant:

After completing this task, evaluate if any learnings should be logged to .learnings/ using the self-improvement skill format.

Or use quick prompts:

• "Log this to learnings"
• "Create a skill from this solution"
• "Check .learnings/ for related issues"