An architectural blueprint and audit tool designed to keep your OpenClaw bootstrap filesystem lean, organized, and optimized for low token usage.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install md-bootstrap-optimizer
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 md-bootstrap-optimizer 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 Bootstrap Optimizer?
Bootstrap Optimizer is a specialized workflow and system design tool built to optimize Openclaw Skills workspaces. As AI agents interact and accumulate guidelines, core files like MEMORY.md and TOOLS.md naturally bloat. This results in truncated rules, lost reasoning capacity, and wasted LLM context. This skill addresses this exact bottleneck by introducing a clean, structured design paradigm.
By leveraging an intelligent three-layer architecture (Core, Execution, and Archive), this system separates permanent atomic principles from on-demand task instructions. This ensures your agent loads only what it needs, when it needs it, keeping your system fast, reliable, and context-efficient.
Bootstrap Optimizer Use Cases
- Your AI agent is experiencing rule truncation or forgetting critical system guidelines due to context bloat.
- You want to structure multi-agent memory files using best practices for Openclaw Skills.
- Your workspace load times are slow because detailed procedures are globally loaded on every session.
- You need to automate workspace auditing to keep bootstrap files under strict token budgets.
How Bootstrap Optimizer Works
- Core Layer Evaluation: Permanently true atomic principles (such as those in MEMORY.md or AGENTS.md) are consolidated and limited to under 3KB.
- Execution Layer Triggering: Highly detailed task procedures are moved to on-demand files (like SCHEDULING.md or PROJECT.md) and loaded only when specific scenes run.
- Mother-Child Splitting: Heavy indexing files such as TOOLS.md are partitioned into a lightweight mother index (< 3KB) and detailed child files loaded on demand.
- Archive Isolation: Historical insights and maintenance logs are assigned to the Archive layer where they reside on disk but never consume session context.
- Workspace Auditing: Command-line audit checks evaluate your file sizes and flag elements exceeding the target thresholds.
Bootstrap Optimizer Setup
To get started with one of the essential Openclaw Skills for workspace optimization, install it via ClawHub:
npx clawhub@latest install md-bootstrap-optimizer
Alternatively, you can manually clone and copy the skill directory into your local configuration:
cp -r md-bootstrap-optimizer ~/.openclaw/skills/
Once installed, map your workspace directory path and configure your desired file-size thresholds.
Bootstrap Optimizer Data Schema & Taxonomy
The layout of the optimized workspace budget follows a strict hierarchy designed for Openclaw Skills efficiency:
| File Type | Target Max Size | Loading Status | Core Purpose |
|---|---|---|---|
| MEMORY.md | < 3KB | Loaded Every Session | Permanent atomic principles, not daily facts |
| AGENTS.md | < 5KB | Loaded Every Session | Core operational procedures and identity anchors |
| TOOLS.md (Mother) | < 3KB | Loaded Every Session | Index pointing to specialized skill sets |
| Execution Files | No strict limit | Loaded on demand | Task-specific procedures and detailed context |
| Archive Files | No limit | Never loaded | Reference material, history logs, and past insights |
Directory Layout
~/.openclaw/workspace/- Houses the managed core Markdown files.~/.openclaw/skills/- Custom and installed modules including the optimizer.
Bootstrap Optimizer Advanced Features
- Mother-Child Pattern Routing: Automate splitting complex rule lists into lightweight, index-based master files and dynamically resolved child records.
- Fact-Principle Separation: Programmatic segregation patterns that keep operational facts out of bootstrap and store them in transient session logs.
- Automated Workspace Diagnostics: Shell-based threshold monitoring script that acts as an alert mechanism before your agent exceeds active context budgets.
- Multi-Agent Protocol Integration: Ready-to-use template structure compatible with other highly optimized Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Raspberry Pi I2C Sensor Development Guide
An advanced guide and toolset for developing, configuring, and optimizing Raspberry Pi I2C sensors like the BMI270 IMU and MAX30205 temperature sensor.
Trading Agents for Futures
A zero-API, rule-based six-dimensional analysis and decision engine for commodities futures markets, designed to power both AI agents and human traders.
iOS SMSSDK Integration
An automated workflow skill designed to smoothly integrate MobTech SMSSDK into iOS projects with minimal disruption.
iFlow Search
Equip your AI agents with real-time web search, image discovery, and LLM-friendly webpage extraction capabilities using the iFlow Search API.
Pet Toy Interaction Activity Analysis
An AI behavioral tracking tool that uses computer vision to analyze pet play area video feeds, quantify toy interactions, and generate daily health and wellness activity curves.
Xiangyun Passport OCR
An AI agent skill powered by the Xiangyun API to perform highly accurate, structured passport OCR and metadata extraction from images.
Featured*
