self-improving-agent

name	self-improving-agent
description	Instinct-based continuous learning system. Captures atomic learnings (instincts) with confidence scoring, supports project-scoped vs global scope, and evolves instincts into skills/commands/agents. Use when: (1) A command fails, (2) User corrects you, (3) Discovering patterns, (4) Need to review or evolve learned behaviors. Supports both v1 (markdown-based) and v2 (instinct-based) modes.
metadata	{"version":"2.1","origin":"ECC + OpenClaw"}

Self-Improving Agent Skill

An advanced learning system that turns Claude Code sessions into reusable knowledge through atomic "instincts" - small learned behaviors with confidence scoring and project scope isolation.

v2.1 adds project-scoped instincts — React patterns stay in your React project, Python conventions stay in your Python project, and universal patterns are shared globally.

Quick Reference

Situation	Action
Command/operation fails	Log instinct or v1 learning
User corrects you	Create instinct with correction trigger
Discovering patterns	Log instinct with confidence score
Review learned behaviors	/instinct-status
Evolve instincts to skills	/evolve
Promote project → global	/promote
Setup observation hooks	Enable PreToolUse/PostToolUse hooks

Two Learning Modes

Mode 1: Instinct-Based (v2) - RECOMMENDED

Atomic, confidence-weighted behaviors with project isolation:

---
id: prefer-functional-style
trigger: "when writing new functions"
confidence: 0.7
domain: "code-style"
scope: project
project_id: "a1b2c3d4e5f6"
---

# Prefer Functional Style

## Action
Use functional patterns over classes when appropriate.

## Evidence
- Observed 5 instances of functional pattern preference
- User corrected class-based approach on 2025-01-15

Mode 2: Markdown-Based (v1) - LEGACY

Traditional learning entries for complex, narrative learnings:

## [LRN-YYYYMMDD-XXX] category
**Priority**: high | **Status**: pending | **Area**: backend

### Summary
Detailed description of what was learned

### Details
Full context and explanation

Use v2 (instincts) for behavioral patterns, v1 (markdown) for complex incident analysis.

---

Instinct-Based Learning (v2)

The Instinct Model

An instinct is a small, atomic learned behavior:

Properties:

• Atomic — one trigger, one action
• Confidence-weighted — 0.3 = tentative, 0.9 = near certain
• Domain-tagged — code-style, testing, git, debugging, workflow, security, etc.
• Evidence-backed — tracks what observations created it
• Scope-aware — project (default) or global

Confidence Scoring

Score	Meaning	Behavior
0.3	Tentative	Suggested but not enforced
0.5	Moderate	Applied when relevant
0.7	Strong	Auto-approved for application
0.9	Near-certain	Core behavior

Confidence increases when:

• Pattern is repeatedly observed
• User doesn't correct the suggested behavior
• Similar instincts from other sources agree

Confidence decreases when:

• User explicitly corrects the behavior
• Pattern isn't observed for extended periods
• Contradicting evidence appears

Scope Decision Guide

Pattern Type	Scope	Examples
Language/framework conventions	project	"Use React hooks", "Follow Django REST patterns"
File structure preferences	project	"Tests in __tests__/", "Components in src/components/"
Code style	project	"Use functional style", "Prefer dataclasses"
Security practices	global	"Validate user input", "Sanitize SQL"
General best practices	global	"Write tests first", "Always handle errors"
Tool workflow preferences	global	"Grep before Edit", "Read before Write"
Git practices	global	"Conventional commits", "Small focused commits"

Project Detection

The system automatically detects your current project:

1. CLAUDE_PROJECT_DIR env var (highest priority)
2. git remote get-url origin — hashed to create a portable project ID
3. git rev-parse --show-toplevel — fallback using repo path
4. Global fallback — if no project detected, instincts go to global scope

Each project gets a 12-character hash ID (e.g., a1b2c3d4e5f6).

v2 Commands

Command	Description
/instinct-status	Show all instincts (project-scoped + global) with confidence
/evolve	Cluster related instincts into skills/commands, suggest promotions
/instinct-export	Export instincts (filterable by scope/domain)
/instinct-import <file>	Import instincts with scope control
/promote [id]	Promote project instincts to global scope
/projects	List all known projects and their instinct counts

/instinct-status Example

Project: my-react-app (a1b2c3d4e5f6)
├─ prefer-functional-style.yaml (0.7) [project]
├─ use-react-hooks.yaml (0.9) [project]
└─ jest-testing-patterns.yaml (0.6) [project]

Global Instincts:
├─ always-validate-input.yaml (0.85) [global]
├─ grep-before-edit.yaml (0.6) [global]
└─ conventional-commits.yaml (0.75) [global]

/evolve Workflow

Clusters related instincts and generates:

• Skills — domain-specific workflows
• Commands — slash commands for common tasks
• Agents — specialized sub-agents

/evolve
# Analyzes instincts and suggests:
# - "Create skill: react-testing-workflow.md"
# - "Create command: /test-component"
# - "Promote prefer-functional-style to global (seen in 3 projects)"

/promote Workflow

Promote project-scoped instincts to global when proven across projects:

/promote prefer-explicit-errors
# Promotes the instinct from current project to global scope

Auto-promotion criteria:

• Same instinct ID in 2+ projects
• Average confidence >= 0.8

File Structure (v2)

~/.claude/homunculus/
├── identity.json           # Your profile, technical level
├── projects.json           # Registry: project hash → name/path/remote
├── observations.jsonl      # Global observations (fallback)
├── instincts/
│   ├── personal/           # Global auto-learned instincts
│   └── inherited/          # Global imported instincts
├── evolved/
│   ├── agents/             # Global generated agents
│   ├── skills/             # Global generated skills
│   └── commands/           # Global generated commands
└── projects/
    ├── a1b2c3d4e5f6/       # Project hash
    │   ├── observations.jsonl
    │   ├── observations.archive/
    │   ├── instincts/
    │   │   ├── personal/   # Project-specific auto-learned
    │   │   └── inherited/  # Project-specific imported
    │   └── evolved/
    │       ├── skills/
    │       ├── commands/
    │       └── agents/
    └── f6e5d4c3b2a1/       # Another project

Enabling Observation Hooks (v2)

Add to your ~/.claude/settings.json:

{
  "hooks": {
    "PreToolUse": [{
      "matcher": "*",
      "hooks": [{
        "type": "command",
        "command": "~/.claude/skills/self-improving-agent/hooks/observe.sh"
      }]
    }],
    "PostToolUse": [{
      "matcher": "*",
      "hooks": [{
        "type": "command",
        "command": "~/.claude/skills/self-improving-agent/hooks/observe.sh"
      }]
    }]
  }
}

Why hooks? Hooks fire 100% of the time, deterministically. Skills fire ~50-80% based on Claude's judgment.

---

OpenClaw is the primary platform for this skill. It uses workspace-based prompt injection with automatic skill loading.

Installation

Via ClawdHub (recommended):

clawdhub install self-improving-agent

Manual:

git clone https://github.com/peterskoett/self-improving-agent.git ~/.openclaw/skills/self-improving-agent

Remade for openclaw from original repo : https://github.com/pskoett/pskoett-ai-skills - https://github.com/pskoett/pskoett-ai-skills/tree/main/skills/self-improvement

Workspace Structure

OpenClaw injects these files into every session:

~/.openclaw/workspace/
├── AGENTS.md          # Multi-agent workflows, delegation patterns
├── SOUL.md            # Behavioral guidelines, personality, principles
├── TOOLS.md           # Tool capabilities, integration gotchas
├── MEMORY.md          # Long-term memory (main session only)
├── memory/            # Daily memory files
│   └── YYYY-MM-DD.md
└── .learnings/        # This skill's log files
    ├── LEARNINGS.md
    ├── ERRORS.md
    └── FEATURE_REQUESTS.md

Create Learning Files

mkdir -p ~/.openclaw/workspace/.learnings

Then create the log files (or copy from assets/):

• LEARNINGS.md — corrections, knowledge gaps, best practices
• ERRORS.md — command failures, exceptions
• FEATURE_REQUESTS.md — user-requested capabilities

Promotion Targets

When learnings prove broadly applicable, promote them to workspace files:

Learning Type	Promote To	Example
Behavioral patterns	SOUL.md	"Be concise, avoid disclaimers"
Workflow improvements	AGENTS.md	"Spawn sub-agents for long tasks"
Tool gotchas	TOOLS.md	"Git push needs auth configured first"

Inter-Session Communication

OpenClaw provides tools to share learnings across sessions:

• sessions_list — View active/recent sessions
• sessions_history — Read another session's transcript
• sessions_send — Send a learning to another session
• sessions_spawn — Spawn a sub-agent for background work

Optional: Enable Hook

For automatic reminders at session start:

# Copy hook to OpenClaw hooks directory
cp -r hooks/openclaw ~/.openclaw/hooks/self-improvement

# Enable it
openclaw hooks enable self-improvement

See references/openclaw-integration.md for complete details.

---

Generic Setup (Other Agents)

For Claude Code, Codex, Copilot, or other agents, create .learnings/ in your project:

mkdir -p .learnings

Copy templates from assets/ or create files with headers.

Add reference to agent files AGENTS.md, CLAUDE.md, or .github/copilot-instructions.md to remind yourself to log learnings. (this is an alternative to hook-based reminders)

Self-Improvement Workflow

When errors or corrections occur:

1. Log to .learnings/ERRORS.md, LEARNINGS.md, or FEATURE_REQUESTS.md
2. Review and promote broadly applicable learnings to:
   • CLAUDE.md - project facts and conventions
   • AGENTS.md - workflows and automation
   • .github/copilot-instructions.md - Copilot context

Logging Formats

v2: Instinct Format (RECOMMENDED for behavioral patterns)

Create atomic instinct files in ~/.claude/homunculus/instincts/personal/ or project-scoped:

---
id: unique-instinct-id
trigger: "when to apply this instinct"
confidence: 0.7
domain: "code-style|testing|git|debugging|workflow|security|infra"
source: "session-observation|user-correction|pattern-detection"
scope: "project|global"
project_id: "a1b2c3d4e5f6"  # if scope: project
project_name: "my-project"
created_at: "2025-01-15T10:00:00Z"
updated_at: "2025-01-15T10:00:00Z"
evidence_count: 3
---

# Instinct Title

## Action
What to do when triggered.

## Rationale
Why this behavior is preferred.

## Examples

### Positive
```typescript
// Good example

Negative

// Bad example

Evidence

• Observed 3 instances of this pattern
• User corrected opposite approach on 2025-01-10

**File naming:** `~/.claude/homunculus/instincts/personal/{instinct-id}.yaml`

### v1: Markdown Format (for complex learnings)

#### Learning Entry

Append to `.learnings/LEARNINGS.md`:

```markdown
## [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 | simplify-and-harden
- Related Files: path/to/file.ext
- Tags: tag1, tag2
- See Also: LRN-20250110-001
- Pattern-Key: simplify.dead_code | harden.input_validation

---

v2 vs v1 Comparison

Feature	v1 (Markdown)	v2 (Instincts)
Granularity	Full skills	Atomic "instincts"
Confidence	None	0.3-0.9 weighted
Scope	Global only	Project-scoped + global
Observation	Stop hook (session end)	PreToolUse/PostToolUse (100% reliable)
Analysis	Main context	Background agent (Haiku)
Evolution	Direct to skill	Instincts → cluster → skill/command/agent
Sharing	None	Export/import instincts
Best for	Complex incidents	Behavioral patterns

Migration from v1 to v2

For existing v1 users: v2 is fully backward compatible:

• Existing global instincts still work
• Existing .learnings/*.md files still work
• Gradual migration: run both in parallel

Recommended approach:

1. Start using v2 instincts for new behavioral patterns
2. Keep v1 markdown for complex incident analysis
3. Use /evolve to convert related v1 learnings into v2 instincts
4. Promote high-confidence instincts to skills

---

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

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

Target	What Belongs There
CLAUDE.md	Project facts, conventions, gotchas for all Claude interactions
AGENTS.md	Agent-specific workflows, tool usage patterns, automation rules
.github/copilot-instructions.md	Project context and conventions for GitHub Copilot
SOUL.md	Behavioral guidelines, communication style, principles (OpenClaw workspace)
TOOLS.md	Tool capabilities, usage patterns, integration gotchas (OpenClaw workspace)

How to Promote

1. Distill the learning into a concise rule or fact
2. Add to appropriate section in target file (create file if needed)
3. Update original entry:
   • Change **Status**: pending → **Status**: promoted
   • Add **Promoted**: CLAUDE.md, AGENTS.md, or .github/copilot-instructions.md

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
• SOUL.md / TOOLS.md for OpenClaw workspace-level guidance when applicable

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

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

Priority	When to Use
critical	Blocks core functionality, data loss risk, security issue
high	Significant impact, affects common workflows, recurring issue
medium	Moderate impact, workaround exists
low	Minor inconvenience, edge case, nice-to-have

Area Tags

Use to filter learnings by codebase region:

Area	Scope
frontend	UI, components, client-side code
backend	API, services, server-side code
infra	CI/CD, deployment, Docker, cloud
tests	Test files, testing utilities, coverage
docs	Documentation, comments, READMEs
config	Configuration files, environment, settings

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 aggressively - if in doubt, add to CLAUDE.md or .github/copilot-instructions.md
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

Enable automatic reminders through agent hooks. This is opt-in - you must explicitly configure hooks.

Quick Setup (Claude Code / Codex)

Create .claude/settings.json in your project:

{
  "hooks": {
    "UserPromptSubmit": [{
      "matcher": "",
      "hooks": [{
        "type": "command",
        "command": "./skills/self-improvement/scripts/activator.sh"
      }]
    }]
  }
}

This injects a learning evaluation reminder after each prompt (~50-100 tokens overhead).

Full Setup (With Error Detection)

{
  "hooks": {
    "UserPromptSubmit": [{
      "matcher": "",
      "hooks": [{
        "type": "command",
        "command": "./skills/self-improvement/scripts/activator.sh"
      }]
    }],
    "PostToolUse": [{
      "matcher": "Bash",
      "hooks": [{
        "type": "command",
        "command": "./skills/self-improvement/scripts/error-detector.sh"
      }]
    }]
  }
}

Available Hook Scripts

Script	Hook Type	Purpose
scripts/activator.sh	UserPromptSubmit	Reminds to evaluate learnings after tasks
scripts/error-detector.sh	PostToolUse (Bash)	Triggers on command errors

See references/hooks-setup.md for detailed configuration and troubleshooting.

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:

Criterion	Description
Recurring	Has See Also links to 2+ similar issues
Verified	Status is resolved with working fix
Non-obvious	Required actual debugging/investigation to discover
Broadly applicable	Not project-specific; useful across codebases
User-flagged	User says "save this as a skill" or similar

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
   

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

This skill works across different AI coding agents with agent-specific activation.

Claude Code

Activation: Hooks (UserPromptSubmit, PostToolUse) Setup: .claude/settings.json with hook configuration Detection: Automatic via hook scripts

Codex CLI

Activation: Hooks (same pattern as Claude Code) Setup: .codex/settings.json with hook configuration Detection: Automatic via hook scripts

GitHub Copilot

Activation: Manual (no hook support) Setup: Add to .github/copilot-instructions.md:

## Self-Improvement

After solving non-obvious issues, consider logging to `.learnings/`:
1. Use format from self-improvement skill
2. Link related entries with See Also
3. Promote high-value learnings to skills

Ask in chat: "Should I log this as a learning?"

Detection: Manual review at session end

OpenClaw

Activation: Workspace injection + inter-agent messaging Setup: See "OpenClaw Setup" section above Detection: Via session tools and workspace files

Agent-Agnostic Guidance

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"
• "/instinct-status" (v2)
• "/evolve" (v2)

---

Privacy

• Observations stay local on your machine
• Project-scoped instincts are isolated per project
• Only instincts (patterns) can be exported — not raw observations
• No actual code or conversation content is shared
• You control what gets exported and promoted

References

Resource	Description
everything-claude-code	ECC project that inspired v2 instinct-based architecture
Homunculus	Community project that influenced v2 design
OpenClaw	Workspace-based multi-agent platform
Agent Skills Spec	https://agentskills.io/specification

---

Instinct-based learning: teaching Claude your patterns, one project at a time.