
Vectorless RAG engine that finds answers by reasoning over a document's page tree
- Open source
- Open-source agent
- 40.5K GitHub stars
- Files and documents

Vectorless RAG engine that finds answers by reasoning over a document's page tree

Open-source knowledge framework for RAG Q&A, agent tasks and auto-built wikis

Self-hosted file manager in one Go binary with a built-in MCP server for agents
Not checked yetConverts invoice PDFs to EN 16931 e-invoices (ZUGFeRD, Factur-X, XRechnung) locally
Runs locallyEdit Markdown or GEML files one section at a time, refusing writes that break the file
Not checked yetFills Korean workers' compensation forms as PDF or HWPX through chat
Not checked yet
Git-backed content governance for AI agents, with 23 deterministic MCP tools
Not checked yetTurn any web page into clean Markdown with one tool, run locally with no API key
Runs locallyCheck AI-written documents against versioned contracts before release, with audit evidence
Runs locallyConvert, validate and inspect FIT, TCX, ZWO and Garmin workout files from an agent
Not checked yetSearch and read AtoM archival catalogues, and propose edits for archivists to review
Not checked yetRead Chinese OFD documents: e-invoice (fapiao) fields, full text and file metadata
Runs locallyTranscribe local audio files and public URLs from your editor through SpeechWeave
Runs locallyFind the right free in-browser file tool and run small text helpers, no uploads
Not checked yetEncrypted peer-to-peer file transfer and chat for people and AI agents, no accounts
Not checked yetA deterministic compiler that turns a course spec into an LMS-ready SCORM package
Not checked yetRemove duplicate lines or CSV and TSV rows from text, processed per call and never stored
Not checked yet