Name: self-improving-agent
Author: ziyabicilecommerce-hub

Prompt file imported from ziyabicilecommerce-hub/ai-cash-machine (.claude/commands/self-improving-agent.md). Fill in {{arguments}} before use. Copyright stays with the author.

description: Curate Claude Code's auto-memory into durable project knowledge. Analyze MEMORY.md for patterns, promote proven learnings to CLAUDE.md and .claude/rules/, extract recurring solutions into reusable skills. Use when: (1) reviewing what Claude has learned about your project, (2) graduating a pattern fr

Self-Improving Agent

Auto-memory captures. This plugin curates.

Claude Code's auto-memory (v2.1.32+) automatically records project patterns, debugging insights, and your preferences in MEMORY.md. This plugin adds the intelligence layer: it analyzes what Claude has learned, promotes proven patterns into project rules, and extracts recurring solutions into reusable skills.

Quick Reference

Command	What it does
`/si:memory-review`	Analyze MEMORY.md — find promotion candidates, stale entries, consolidation opportunities
`/si:promote`	Graduate a pattern from MEMORY.md → CLAUDE.md or `.claude/rules/`
`/si:extract`	Turn a proven pattern into a standalone skill
`/si:memory-status`	Memory health dashboard — line counts, topic files, recommendations
`/si:remember`	Explicitly save important knowledge to auto-memory

How It Fits Together

┌─────────────────────────────────────────────────────────┐
│                  Claude Code Memory Stack                │
├─────────────┬──────────────────┬────────────────────────┤
│  CLAUDE.md  │   Auto Memory    │   Session Memory       │
│  (you write)│   (Claude writes)│   (Claude writes)      │
│  Rules &    │   MEMORY.md      │   Conversation logs    │
│  standards  │   + topic files  │   + continuity         │
│  Full load  │   First 200 lines│   Contextual load      │
├─────────────┴──────────────────┴────────────────────────┤
│              ↑ /si:promote        ↑ /si:memory-review   │
│         Self-Improving Agent (this plugin)               │
│              ↓ /si:extract    ↓ /si:remember            │
├─────────────────────────────────────────────────────────┤
│  .claude/rules/    │    New Skills    │   Error Logs     │
│  (scoped rules)    │    (extracted)   │   (auto-captured)│
└─────────────────────────────────────────────────────────┘

Installation

Claude Code (Plugin)

/plugin marketplace add alirezarezvani/claude-skills
/plugin install self-improving-agent@claude-code-skills

OpenClaw

clawhub install self-improving-agent

Codex CLI

./scripts/codex-install.sh --skill self-improving-agent

Memory Architecture

Where things live

File	Who writes	Scope	Loaded
`./CLAUDE.md`	You (+ `/si:promote`)	Project rules	Full file, every session
`~/.claude/CLAUDE.md`	You	Global preferences	Full file, every session
`~/.claude/projects/<path>/memory/MEMORY.md`	Claude (auto)	Project learnings	First 200 lines
`~/.claude/projects/<path>/memory/*.md`	Claude (overflow)	Topic-specific notes	On demand
`.claude/rules/*.md`	You (+ `/si:promote`)	Scoped rules	When matching files open

The promotion lifecycle

1. Claude discovers pattern → auto-memory (MEMORY.md)
2. Pattern recurs 2-3x → /si:memory-review flags it as promotion candidate
3. You approve → /si:promote graduates it to CLAUDE.md or rules/
4. Pattern becomes an enforced rule, not just a note
5. MEMORY.md entry removed → frees space for new learnings

Core Concepts

Auto-memory is capture, not curation

Auto-memory is excellent at recording what Claude learns. But it has no judgment about:

Which learnings are temporary vs. permanent
Which patterns should become enforced rules
When the 200-line limit is wasting space on stale entries
Which solutions are good enough to become reusable skills

That's what this plugin does.

Promotion = graduation

When you promote a learning, it moves from Claude's scratchpad (MEMORY.md) to your project's rule system (CLAUDE.md or .claude/rules/). The difference matters:

MEMORY.md: "I noticed this project uses pnpm" (background context)
CLAUDE.md: "Use pnpm, not npm" (enforced instruction)

Promoted rules have higher priority and load in full (not truncated at 200 lines).

Rules directory for scoped knowledge

Not everything belongs in CLAUDE.md. Use .claude/rules/ for patterns that only apply to specific file types:

# .claude/rules/api-testing.md
---
paths:
  - "src/api/**/*.test.ts"
  - "tests/api/**/*"
---
- Use supertest for API endpoint testing
- Mock external services with msw
- Always test error responses, not just happy paths

This loads only when Claude works with API test files — zero overhead otherwise.

Agents

memory-analyst

Analyzes MEMORY.md and topic files to identify:

Entries that recur across sessions (promotion candidates)
Stale entries referencing deleted files or old patterns
Related entries that should be consolidated
Gaps between what MEMORY.md knows and what CLAUDE.md enforces

skill-extractor

Takes a proven pattern and generates a complete skill:

SKILL.md with proper frontmatter
Reference documentation
Examples and edge cases
Ready for /plugin install or clawhub publish

Hooks

error-capture (PostToolUse → Bash)

Monitors command output for errors. When detected, appends a structured entry to auto-memory with:

The command that failed
Error output (truncated)
Timestamp and context
Suggested category

Token overhead: Zero on success. ~30 tokens only when an error is detected.

Platform Support

Platform	Memory System	Plugin Works?
Claude Code	Auto-memory (MEMORY.md)	✅ Full support
OpenClaw	workspace/MEMORY.md	✅ Adapted (reads workspace memory)
Codex CLI	AGENTS.md	✅ Adapted (reads AGENTS.md patterns)
GitHub Copilot	`.github/copilot-instructions.md`	⚠️ Manual promotion only

Claude Code Memory Docs
pskoett/self-improving-agent — inspiration
playwright-pro — sister plugin in this repo

Apply the above self-improving-agent instructions to the user request below.

Request: {{arguments}}

Install self-improving-agent and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

--- description: Curate Claude Code's auto-memory into durable project knowledge. Analyze MEMORY.md for patterns, promote proven learnings to CLAUDE.md and .claude/rules/, extract recurring solutions into reusable skills. Use when: (1) reviewing what Claude has learned about your project, (2) graduating a pattern fr --- # Self-Improving Agent > Auto-memory captures. This plugin curates. Claude Code's auto-memory (v2.1.32+) automatically records project patterns, debugging insights, and your preferences in `MEMORY.md`. This plugin adds the intelligence layer: it analyzes what Claude has learned, promotes proven patterns into project rules, and extracts recurring solutions into reusable skills. ## Quick Reference | Command | What it does | |---------|-------------| | `/si:memory-review` | Analyze MEMORY.md — find promotion candidates, stale entries, consolidation opportunities | | `/si:promote` | Graduate a pattern from MEMORY.md → CLAUDE.md or `.claude/rules/` | | `/si:extract` | Turn a proven pattern into a standalone skill | | `/si:memory-status` | Memory health dashboard — line counts, topic files, recommendations | | `/si:remember` | Explicitly save important knowledge to auto-memory | ## How It Fits Together ``` ┌─────────────────────────────────────────────────────────┐ │ Claude Code Memory Stack │ ├─────────────┬──────────────────┬────────────────────────┤ │ CLAUDE.md │ Auto Memory │ Session Memory │ │ (you write)│ (Claude writes)│ (Claude writes) │ │ Rules & │ MEMORY.md │ Conversation logs │ │ standards │ + topic files │ + continuity │ │ Full load │ First 200 lines│ Contextual load │ ├─────────────┴──────────────────┴────────────────────────┤ │ ↑ /si:promote ↑ /si:memory-review │ │ Self-Improving Agent (this plugin) │ │ ↓ /si:extract ↓ /si:remember │ ├─────────────────────────────────────────────────────────┤ │ .claude/rules/ │ New Skills │ Error Logs │ │ (scoped rules) │ (extracted) │ (auto-captured)│ └─────────────────────────────────────────────────────────┘ ``` ## Installation ### Claude Code (Plugin) ``` /plugin marketplace add alirezarezvani/claude-skills /plugin install self-improving-agent@claude-code-skills ``` ### OpenClaw ```bash clawhub install self-improving-agent ``` ### Codex CLI ```bash ./scripts/codex-install.sh --skill self-improving-agent ``` ## Memory Architecture ### Where things live | File | Who writes | Scope | Loaded | |------|-----------|-------|--------| | `./CLAUDE.md` | You (+ `/si:promote`) | Project rules | Full file, every session | | `~/.claude/CLAUDE.md` | You | Global preferences | Full file, every session | | `~/.claude/projects/<path>/memory/MEMORY.md` | Claude (auto) | Project learnings | First 200 lines | | `~/.claude/projects/<path>/memory/*.md` | Claude (overflow) | Topic-specific notes | On demand | | `.claude/rules/*.md` | You (+ `/si:promote`) | Scoped rules | When matching files open | ### The promotion lifecycle ``` 1. Claude discovers pattern → auto-memory (MEMORY.md) 2. Pattern recurs 2-3x → /si:memory-review flags it as promotion candidate 3. You approve → /si:promote graduates it to CLAUDE.md or rules/ 4. Pattern becomes an enforced rule, not just a note 5. MEMORY.md entry removed → frees space for new learnings ``` ## Core Concepts ### Auto-memory is capture, not curation Auto-memory is excellent at recording what Claude learns. But it has no judgment about: - Which learnings are temporary vs. permanent - Which patterns should become enforced rules - When the 200-line limit is wasting space on stale entries - Which solutions are good enough to become reusable skills That's what this plugin does. ### Promotion = graduation When you promote a learning, it moves from Claude's scratchpad (MEMORY.md) to your project's rule system (CLAUDE.md or `.claude/rules/`). The difference matters: - **MEMORY.md**: "I noticed this project uses pnpm" (background context) - **CLAUDE.md**: "Use pnpm, not npm" (enforced instruction) Promoted rules have higher priority and load in full (not truncated at 200 lines). ### Rules directory for scoped knowledge Not everything belongs in CLAUDE.md. Use `.claude/rules/` for patterns that only apply to specific file types: ```yaml # .claude/rules/api-testing.md --- paths: - "src/api/**/*.test.ts" - "tests/api/**/*" --- - Use supertest for API endpoint testing - Mock external services with msw - Always test error responses, not just happy paths ``` This loads only when Claude works with API test files — zero overhead otherwise. ## Agents ### memory-analyst Analyzes MEMORY.md and topic files to identify: - Entries that recur across sessions (promotion candidates) - Stale entries referencing deleted files or old patterns - Related entries that should be consolidated - Gaps between what MEMORY.md knows and what CLAUDE.md enforces ### skill-extractor Takes a proven pattern and generates a complete skill: - SKILL.md with proper frontmatter - Reference documentation - Examples and edge cases - Ready for `/plugin install` or `clawhub publish` ## Hooks ### error-capture (PostToolUse → Bash) Monitors command output for errors. When detected, appends a structured entry to auto-memory with: - The command that failed - Error output (truncated) - Timestamp and context - Suggested category **Token overhead:** Zero on success. ~30 tokens only when an error is detected. ## Platform Support | Platform | Memory System | Plugin Works? | |----------|--------------|---------------| | Claude Code | Auto-memory (MEMORY.md) | ✅ Full support | | OpenClaw | workspace/MEMORY.md | ✅ Adapted (reads workspace memory) | | Codex CLI | AGENTS.md | ✅ Adapted (reads AGENTS.md patterns) | | GitHub Copilot | `.github/copilot-instructions.md` | ⚠️ Manual promotion only | ## Related - [Claude Code Memory Docs](https://code.claude.com/docs/en/memory) - [pskoett/self-improving-agent](https://clawhub.ai/pskoett/self-improving-agent) — inspiration - [playwright-pro](engineering-team/playwright-pro/) — sister plugin in this repo --- Apply the above **self-improving-agent** instructions to the user request below. Request: {{arguments}}

{ "prompt": "---\ndescription: Curate Claude Code's auto-memory into durable project knowledge. Analyze MEMORY.md for patterns, promote proven learnings to CLAUDE.md and .claude/rules/, extract recurring solutions into reusable skills. Use when: (1) reviewing what Claude has learned about your project, (2) graduating a pattern fr\n---\n\n\n# Self-Improving Agent\n\n> Auto-memory captures. This plugin curates.\n\nClaude Code's auto-memory (v2.1.32+) automatically records project patterns, debugging insights, and your preferences in `MEMORY.md`. This plugin adds the intelligence layer: it analyzes what Claude has learned, promotes proven patterns into project rules, and extracts recurring solutions into reusable skills.\n\n## Quick Reference\n\n| Command | What it does |\n|---------|-------------|\n| `/si:memory-review` | Analyze MEMORY.md — find promotion candidates, stale entries, consolidation opportunities |\n| `/si:promote` | Graduate a pattern from MEMORY.md → CLAUDE.md or `.claude/rules/` |\n| `/si:extract` | Turn a proven pattern into a standalone skill |\n| `/si:memory-status` | Memory health dashboard — line counts, topic files, recommendations |\n| `/si:remember` | Explicitly save important knowledge to auto-memory |\n\n## How It Fits Together\n\n```\n┌─────────────────────────────────────────────────────────┐\n│ Claude Code Memory Stack │\n├─────────────┬──────────────────┬────────────────────────┤\n│ CLAUDE.md │ Auto Memory │ Session Memory │\n│ (you write)│ (Claude writes)│ (Claude writes) │\n│ Rules & │ MEMORY.md │ Conversation logs │\n│ standards │ + topic files │ + continuity │\n│ Full load │ First 200 lines│ Contextual load │\n├─────────────┴──────────────────┴────────────────────────┤\n│ ↑ /si:promote ↑ /si:memory-review │\n│ Self-Improving Agent (this plugin) │\n│ ↓ /si:extract ↓ /si:remember │\n├─────────────────────────────────────────────────────────┤\n│ .claude/rules/ │ New Skills │ Error Logs │\n│ (scoped rules) │ (extracted) │ (auto-captured)│\n└─────────────────────────────────────────────────────────┘\n```\n\n## Installation\n\n### Claude Code (Plugin)\n```\n/plugin marketplace add alirezarezvani/claude-skills\n/plugin install self-improving-agent@claude-code-skills\n```\n\n### OpenClaw\n```bash\nclawhub install self-improving-agent\n```\n\n### Codex CLI\n```bash\n./scripts/codex-install.sh --skill self-improving-agent\n```\n\n## Memory Architecture\n\n### Where things live\n\n| File | Who writes | Scope | Loaded |\n|------|-----------|-------|--------|\n| `./CLAUDE.md` | You (+ `/si:promote`) | Project rules | Full file, every session |\n| `~/.claude/CLAUDE.md` | You | Global preferences | Full file, every session |\n| `~/.claude/projects/<path>/memory/MEMORY.md` | Claude (auto) | Project learnings | First 200 lines |\n| `~/.claude/projects/<path>/memory/*.md` | Claude (overflow) | Topic-specific notes | On demand |\n| `.claude/rules/*.md` | You (+ `/si:promote`) | Scoped rules | When matching files open |\n\n### The promotion lifecycle\n\n```\n1. Claude discovers pattern → auto-memory (MEMORY.md)\n2. Pattern recurs 2-3x → /si:memory-review flags it as promotion candidate\n3. You approve → /si:promote graduates it to CLAUDE.md or rules/\n4. Pattern becomes an enforced rule, not just a note\n5. MEMORY.md entry removed → frees space for new learnings\n```\n\n## Core Concepts\n\n### Auto-memory is capture, not curation\n\nAuto-memory is excellent at recording what Claude learns. But it has no judgment about:\n- Which learnings are temporary vs. permanent\n- Which patterns should become enforced rules\n- When the 200-line limit is wasting space on stale entries\n- Which solutions are good enough to become reusable skills\n\nThat's what this plugin does.\n\n### Promotion = graduation\n\nWhen you promote a learning, it moves from Claude's scratchpad (MEMORY.md) to your project's rule system (CLAUDE.md or `.claude/rules/`). The difference matters:\n\n- **MEMORY.md**: \"I noticed this project uses pnpm\" (background context)\n- **CLAUDE.md**: \"Use pnpm, not npm\" (enforced instruction)\n\nPromoted rules have higher priority and load in full (not truncated at 200 lines).\n\n### Rules directory for scoped knowledge\n\nNot everything belongs in CLAUDE.md. Use `.claude/rules/` for patterns that only apply to specific file types:\n\n```yaml\n# .claude/rules/api-testing.md\n---\npaths:\n - \"src/api/**/*.test.ts\"\n - \"tests/api/**/*\"\n---\n- Use supertest for API endpoint testing\n- Mock external services with msw\n- Always test error responses, not just happy paths\n```\n\nThis loads only when Claude works with API test files — zero overhead otherwise.\n\n## Agents\n\n### memory-analyst\nAnalyzes MEMORY.md and topic files to identify:\n- Entries that recur across sessions (promotion candidates)\n- Stale entries referencing deleted files or old patterns\n- Related entries that should be consolidated\n- Gaps between what MEMORY.md knows and what CLAUDE.md enforces\n\n### skill-extractor\nTakes a proven pattern and generates a complete skill:\n- SKILL.md with proper frontmatter\n- Reference documentation\n- Examples and edge cases\n- Ready for `/plugin install` or `clawhub publish`\n\n## Hooks\n\n### error-capture (PostToolUse → Bash)\nMonitors command output for errors. When detected, appends a structured entry to auto-memory with:\n- The command that failed\n- Error output (truncated)\n- Timestamp and context\n- Suggested category\n\n**Token overhead:** Zero on success. ~30 tokens only when an error is detected.\n\n## Platform Support\n\n| Platform | Memory System | Plugin Works? |\n|----------|--------------|---------------|\n| Claude Code | Auto-memory (MEMORY.md) | ✅ Full support |\n| OpenClaw | workspace/MEMORY.md | ✅ Adapted (reads workspace memory) |\n| Codex CLI | AGENTS.md | ✅ Adapted (reads AGENTS.md patterns) |\n| GitHub Copilot | `.github/copilot-instructions.md` | ⚠️ Manual promotion only |\n\n## Related\n\n- [Claude Code Memory Docs](https://code.claude.com/docs/en/memory)\n- [pskoett/self-improving-agent](https://clawhub.ai/pskoett/self-improving-agent) — inspiration\n- [playwright-pro](engineering-team/playwright-pro/) — sister plugin in this repo\n\n---\n\nApply the above **self-improving-agent** instructions to the user request below.\n\nRequest: {{arguments}}", "variables": [ "arguments" ], "notes": "Claude Code's auto-memory (v2.1.32+) automatically records project patterns, debugging insights, and your preferences in `MEMORY.md`. This plugin adds the intelligence layer: it analyzes what Claude has learned, promotes proven patterns into project rules, and extracts recurring solutions into reusable skills.", "metadata": { "source": { "provider": "github-claude-commands", "repository": "https://github.com/ziyabicilecommerce-hub/ai-cash-machine", "path": ".claude/commands/self-improving-agent.md", "ref": "e1e952ca4cb4f55e7ef39152b41fef1c07feee90", "url": "https://github.com/ziyabicilecommerce-hub/ai-cash-machine/blob/e1e952ca4cb4f55e7ef39152b41fef1c07feee90/.claude/commands/self-improving-agent.md", "key": "ziyabicilecommerce-hub/ai-cash-machine/.claude/commands/self-improving-agent.md" } } }

self-improving-agent

About