A specialized skill for syncing repository documentation with actual code state using a routed progressive-disclosure workflow.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install docs-refresh
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install docs-refresh using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
Docs Refresh is a technical utility designed for Openclaw Skills to bridge the gap between codebase evolution and durable documentation. It ensures that your versioned docs remain the authoritative record system by analyzing the current repository state—including git diffs, schema changes, and architecture shifts—and updating the minimal necessary documentation files. By following a platform-neutral approach, it avoids chat-based silos and keeps knowledge close to the code where it belongs. By using Openclaw Skills like this, developers can ensure that technical knowledge is versioned alongside the code it describes.
To integrate this skill into your workflow for Openclaw Skills, ensure the skill directory is properly resolved within your repository context.
# Ensure the skill files are present in your workspace
# The skill will automatically attempt to run the bundled collector
# No external dependencies are required beyond the Openclaw Skills runtime
Docs Refresh organizes documentation updates based on a taxonomy of repository maturity. The following table describes how this Openclaw Skills utility selects target files:
| Mode | Repo State | Primary Documentation Targets |
|---|---|---|
| Bootstrap | No existing system | README.md, AGENTS.md |
| Minimal | Core docs exist | ARCHITECTURE.md, AGENTS.md, top-level docs |
| Structured | Split domains exist | docs/**/index.md, domain-specific reference pages |
| Repair | Broken navigation | Navigation maps, indexes, and cross-link surfaces |
Loading
A specialized process watchdog that monitors and restores working directories to prevent evolver crashes.

A high-performance local AI inference suite designed to run massive LLMs and creative models across Apple Silicon hardware.

A high-rigor five-step reasoning framework designed to force disconfirmation and produce high-confidence conclusions for complex decision-making tasks.

A comprehensive automation suite for the Lark (Feishu) ecosystem that integrates AI agents with documents, multidimensional tables, and knowledge bases.

A powerful AI skill that transforms natural language descriptions into professional, hand-drawn Excalidraw visual diagrams.

a2a.fun is a coordination harness designed for seamless multi-agent collaboration on shared projects, tasks, and proposals.








































