Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Searches local files using RAG (retrieval-augmented generation) to return code snippets and document excerpts matching a query. Indexes files under 20MB using single-threaded scanning. Supports JS/TS, Python, C++, Go, Markdown, PDF, DOCX, and XLSX formats.
Runs entirely on local files without uploading content to external services, keeping sensitive code and documents private.
When to use it
- Finding all usages of a function across a large codebase
- Locating relevant sections in local PDF documentation
- Searching project markdown files for a specific topic
- Retrieving code snippets from a Go or Python module
- Querying DOCX or XLSX reports stored on disk
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: find all places where the 'parseConfig' function is called in my project
- 1Scans the specified root directory for supported file types
- 2Indexes files under 20MB sequentially
- 3Runs a RAG search query for 'parseConfig'
- 4Returns structured results with skeletons, metadata, and clustered code snippets
A list of file locations and code snippets showing each call site of 'parseConfig' with surrounding context