context-harness
Active·★ 42·AGPL-3.0·Updated 2026-10-15
Context Harness is a local-first framework designed for ingesting and retrieving external knowledge into an AI tool's context. It supports various data sources like files, Git repositories, and S3 buckets, storing them in a local SQLite database for efficient keyword and semantic search.
#Context Management#RAG Framework#Embeddings#Hybrid Search#CLI Tool#Data Ingestion#SQLite#AI Tools Integration
01
Features
01Connector-driven ingestion supporting various sources (filesystem, Git, S3, Lua scripts)
02Multi-format file support including PDF, Word, PowerPoint, and Excel
03Flexible embedding pipeline with local, Ollama, and OpenAI providers
04Hybrid retrieval combining keyword (FTS5) and semantic search with configurable weighting
05MCP server to expose context to AI tools like Cursor and Claude via HTTP
02
Compatibility
Linux
OS
Verified via docs
macOS
OS
Verified via docs
Windows
OS
Verified via docs
Ollama
Embedding Provider
Verified via docs
OpenAI
Embedding Provider
Verified via docs
Cursor
AI Tool Integration
Verified via docs
03
Quick start
1
$ curl -L https://github.com/parallax-labs/context-harness/releases/latest/download/ctx-linux-x86_64.tar.gz | tar xz
2
$ sudo mv ctx /usr/local/bin/
04
Use cases
↳Provide AI tools with searchable local project documentation and knowledge bases.
↳Integrate diverse data sources (Git repos, S3, local files) into a unified context for RAG systems.
↳Enable AI assistants like Cursor or Claude to understand and respond based on private or proprietary enterprise data.
↳Build custom AI agents that require access to continuously updated external information.
05
Alternatives
mcp-local-memory★ 60
A lightweight, powerful local memory server for AI agents supporting text, entities, and relations. Enables persistent codebase understanding and user preference management.
hippo-memory★ 784
Biologically-inspired memory for AI agents. Decay, retrieval strengthening, consolidation. Zero runtime deps, SQLite, MCP. Benchmarked retrieval with an opt-in hosted TypeSafe Jev reranker.
squish-memory★ 12
Local-first memory runtime for AI agents. One command to connect Claude Code, Codex, ChatGPT, MCP workflows, and local agents.
mnemo★ 12
Persistent AI memory with hybrid search and embedded sync. Open, free, unlimited.
claude-mem-lite★ 67
Persistent long-term memory for Claude Code via MCP — captures coding decisions, bugfixes, and context across sessions. Hybrid FTS5 + TF-IDF search with episode batching. Single SQLite DB, no external services. Alternative to claude-mem with 600x lower cost.
remnic★ 228
Open-source memory and context for user-aware agents: scoped memory, provenance, retrieval quality, correction, boundaries, evals, and MCP/HTTP access.
Comments
Log in to leave a comment
No comments yet. Be the first!