Boof
Scanned@chiefsegundo
npx machina-cli add skill @chiefsegundo/boof --openclawBoof 🍑
Local-first document processing: PDF → markdown → RAG index → token-efficient analysis.
Documents stay local. Only relevant chunks go to the LLM. Maximum knowledge absorption, minimum token burn.
Quick Reference
Convert + index a document
bash {SKILL_DIR}/scripts/boof.sh /path/to/document.pdf
Convert with custom collection name
bash {SKILL_DIR}/scripts/boof.sh /path/to/document.pdf --collection my-project
Query indexed content
qmd query "your question" -c collection-name
Core Workflow
-
Boof it: Run
boof.shon a PDF. This converts it to markdown via Marker (local ML, no API) and indexes it into QMD for semantic search. -
Query it: Use
qmd queryto retrieve only the relevant chunks. Send those chunks to the LLM — not the entire document. -
Analyze it: The LLM sees focused, relevant excerpts. No wasted tokens, no lost-in-the-middle problems.
When to Use Each Approach
"Analyze this specific aspect of the paper" → Boof + query (cheapest, most focused)
"Summarize this entire document" → Boof, then read the markdown section by section. Summarize each section individually, then merge summaries. See advanced-usage.md.
"Compare findings across multiple papers" → Boof all papers into one collection, then query across them.
"Find where the paper discusses X" → qmd search "X" -c collection for exact match, qmd query "X" -c collection for semantic match.
Output Location
Converted markdown files are saved to knowledge/boofed/ by default (override with --output-dir).
Setup
If boof.sh reports missing dependencies, see setup-guide.md for installation instructions (Marker + QMD).
Environment
MARKER_ENV— Path to marker-pdf Python venv (default:~/.openclaw/tools/marker-env)QMD_BIN— Path to qmd binary (default:~/.bun/bin/qmd)BOOF_OUTPUT_DIR— Default output directory (default:~/.openclaw/workspace/knowledge/boofed)
Overview
Boof converts PDFs and documents into markdown, then builds a local RAG index for efficient, token-conscious analysis. It keeps source files on your machine, sending only relevant chunks to the LLM to maximize knowledge absorption while minimizing token burn.
How This Skill Works
Boof.sh converts a PDF to markdown using Marker (local ML, no API) and indexes it into QMD for semantic search. You then use qmd query to fetch only the most relevant chunks and feed those excerpts to the LLM for analysis, ensuring focused, efficient processing.
When to Use It
- Read/analyze/summarize a PDF
- Process a document
- Boof a file
- Extract information from papers/decks/NOFOs
- Work with large documents without filling the context window
Quick Start
- Step 1: Run boof.sh /path/to/document.pdf to convert to markdown and index it
- Step 2: Query content with qmd query "your question" -c collection-name
- Step 3: Review retrieved excerpts and send them to the LLM for analysis
Best Practices
- Run boof.sh on the target PDF to generate markdown and build the local index
- Organize projects with --collection and customize output with --output-dir
- Query selectively with qmd query to retrieve only relevant chunks
- Batch-process multiple documents before performing analysis
- Review retrieved excerpts and synthesize results to avoid token waste
Example Use Cases
- Analyze a long research paper locally and extract key findings
- Index a stack of NOFOs to pull deadlines and requirements across documents
- Compare findings across multiple decks by Boof-ing them into a single collection
- Summarize a long report section-by-section and merge the summaries
- Build a local knowledge base for a project by ingesting multiple PDFs
