A specialized tool that uses Chain-of-Density iterations to compress verbose SKILL.md files into terse, structured, and highly efficient documentation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install skill-condenser
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 skill-condenser 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 Skill Condenser?
Skill Condenser is a developer-centric utility designed to refactor verbose documentation into a high-density format. It specifically targets SKILL.md files that have become bloated with prose, converting them into structured Markdown featuring bullet lists, tables, and concise code blocks. By applying refined Chain-of-Density (CoD) logic, it ensures that your Openclaw Skills remain readable for humans while being optimized for AI agent context windows.
This skill is essential for maintaining high-performance agentic workflows where token efficiency is critical. It prioritizes technical entities such as commands, configuration options, and error-fix pairs, stripping away filler phrases without losing the core technical value of the original documentation. This makes it the go-to solution for managing complex Openclaw Skills architectures.
Skill Condenser Use Cases
- Compressing SKILL.md files that exceed the 200-line threshold for better agent parsing.
- Converting descriptive prose into actionable tables and bulleted lists for technical clarity.
- Refactoring large documentation sets (500+ lines) into a main skill file with linked references.
- Standardizing the documentation style across a library of Openclaw Skills.
How Skill Condenser Works
- The skill analyzes the target SKILL.md file to identify the existing structure and key technical entities.
- It performs an initial Structure Extraction pass to convert paragraphs into headers and bullet points based on specific word count targets.
- An Entity Densification pass is executed to combine related facts, remove redundant examples, and migrate options into compact tables.
- If the file remains too large, an optional Final Polish pass is conducted to implement progressive disclosure, moving deep technical details to a separate references directory.
- The final output is generated as a valid SKILL.md file, preserving critical frontmatter and section headers.
Skill Condenser Setup
To integrate this utility into your workflow, ensure your environment supports the CoD iteration pattern. Use the following pattern to begin a condensation task:
# Trigger the condensation process for a specific skill file
skill-condenser --input ./path/to/verbose-skill.md --iterations 2
You can also manually pass segments of your Openclaw Skills to the cod-iteration tool for granular control over the densification levels.
Skill Condenser Data Schema & Taxonomy
The Skill Condenser organizes data using a hierarchical taxonomy that prioritizes structured entities over descriptive text. The resulting documentation follows this organization:
| Component | Formatting Strategy | Purpose |
|---|---|---|
| Metadata | YAML Frontmatter | Preserves author, version, and license data exactly. |
| Commands | Triple-backtick Code Blocks | Ensures CLI examples are copy-paste ready. |
| Config Options | Markdown Tables | Compresses multiple parameters into a single view. |
| Troubleshooting | Arrow notation (Error → Fix) | Provides rapid-fire solutions for common issues. |
| Logic Flow | Numbered Lists | Outlines step-by-step processes for Openclaw Skills. |
Skill Condenser Advanced Features
- Skill-format awareness that prevents the destruction of critical Markdown headers.
- Progressive disclosure support for splitting large Openclaw Skills into core and reference modules.
- Entity prioritization logic that keeps high-value code examples while discarding verbose explanations.
- Multi-pass compression targets based on original file length (from 200 to 500+ lines).
- Integration with
cod-iterationfor iterative refinement without losing technical nuance.
SKILL.md
Loading
Related Openclaw Skills
PARA PKM Manager
A powerful organizational skill for managing personal knowledge using the PARA method to optimize for AI agent navigation and human actionability.
Copywriter
A professional-grade skill for generating high-conversion marketing copy, intuitive UX text, and supportive product messaging using proven linguistic frameworks.
Chain-of-Density Summarization
An advanced summarization skill that iteratively injects information into a text summary while maintaining a strict word count.
BAML Code Generation
A specialized skill for generating type-safe BAML code to handle structured LLM outputs, classification, and RAG workflows with 95% compilation success.
VHS Recorder
A specialized automation skill for generating professional terminal recordings and CLI demos using scripted tape files.
apktool
A professional command-line utility for reverse engineering Android APK files, enabling developers to decode, modify, and rebuild applications.
Featured*
