A curation layer for AI agent memory that transforms temporary patterns into permanent project rules and reusable skills.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install auto-memory-pro
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install auto-memory-pro using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is auto-memory-pro?
auto-memory-pro is a specialized extension designed to bridge the gap between passive observation and active rule enforcement for AI agents. While Claude Code's native auto-memory system is excellent at recording project patterns and debugging insights in MEMORY.md, it often lacks the judgment to distinguish between fleeting context and permanent project standards. This tool provides the intelligence layer needed to manage the 200-line memory limit efficiently.
By using these Openclaw Skills, developers can ensure their AI agents operate with high-precision project knowledge. The plugin analyzes what the agent has learned, promotes proven patterns into enforced rules in CLAUDE.md or .claude/rules/, and extracts successful debugging solutions into standalone skills. This prevents memory bloat and ensures that critical project conventions are never truncated or lost, turning the AI agent into a truly self-improving assistant.
auto-memory-pro Use Cases
- Reviewing what an AI agent has learned about a specific project structure or tech stack.
- Graduating a recurring coding pattern from simple notes into strictly enforced project rules.
- Converting a complex debugging solution into a reusable skill for future use.
- Monitoring memory health and capacity to prevent context truncation in large projects.
- Automating the capture of command errors into the agent's long-term project memory.
How auto-memory-pro Works
- The AI agent discovers a pattern during interaction and records it in the auto-memory system (MEMORY.md).
- The user runs the review command to identify recurring patterns and potential promotion candidates.
- Validated patterns are promoted to CLAUDE.md or scoped rule files in the .claude/rules/ directory.
- The plugin removes the original entry from MEMORY.md to free up space for new project learnings.
- Recurring solutions can be processed by the skill-extractor to create portable Openclaw Skills documentation.
auto-memory-pro Setup
To install the plugin for Claude Code, run the following commands:
/plugin marketplace add alirezarezvani/claude-skills
/plugin install self-improving-agent@claude-code-skills
For users leveraging Openclaw Skills in a hub environment, use:
clawhub install self-improving-agent
For Codex CLI users:
./scripts/codex-install.sh --skill self-improving-agent
auto-memory-pro Data Schema & Taxonomy
The skill organizes data across several key files to maintain a hierarchy of knowledge:
| File | Purpose | Scope |
|---|---|---|
| MEMORY.md | Auto-captured learnings | Project-specific (first 200 lines loaded) |
| CLAUDE.md | Enforced project rules | Full file loaded every session |
| .claude/rules/*.md | Scoped instructions | Loaded when matching specific file paths |
| SKILL.md | Reusable logic | Portable documentation and frontmatter for new skills |
auto-memory-pro Advanced Features
- Automated error-capture hooks that record failed command context with zero overhead on success.
- Integrated memory-analyst agent for identifying gaps between current memory and project rules.
- Skill-extractor capability for generating complete SKILL.md files from proven memory patterns.
- Support for scoped rules via the .claude/rules/ directory to minimize token usage.
- Multi-platform compatibility across Claude Code, OpenClaw, and Codex CLI.
SKILL.md
Loading
Related Openclaw Skills
API Design Reviewer
Automate RESTful API analysis, linting, and breaking change detection to ensure high-quality engineering standards.
Whisper Context
A context optimization and long-term memory skill that reduces LLM token usage via delta compression and intelligent caching.
usewhisper-autohook
An automated memory management tool for OpenClaw agents that handles context retrieval and conversation ingestion via Whisper Context.
Gemini Voice Assistant
A voice-to-voice AI assistant leveraging the Gemini Live API for natural, real-time spoken interactions.
Board Deck Builder
A specialized AI agent skill designed to synthesize C-suite insights into high-impact board decks and investor updates.
Board Meeting Protocol
A multi-agent orchestration framework that facilitates isolated, structured C-suite deliberations to drive unbiased executive decisions.
Featured*
