Local First Python Packages

loro

Python bindings for Loro CRDTs

logosdb

LogosDB is a fast semantic vector database for RAG written in C/C++ that provides approximate nearest-neighbor search over embedding vectors with associated text metadata.

basic-memory

AI conversations that actually remember. Never re-explain your project to your AI again. Join our Discord: https://discord.gg/tyvKNccgqN

roam-code

Local codebase intelligence CLI + MCP server for AI coding agents: SQLite code graph, 28 languages, 238 commands, 224 MCP tools, change-safety gates, audit evidence, zero API keys.

mind-mem

Persistent AI memory for Claude Code, OpenClaw, and any MCP-compatible agent. v4.0.9 — Predicate.register() runtime API + CI fully green (26/26 jobs, 12 OS×Python matrix rows). BM25F+vector hybrid, governance-aware, 5428+ tests, 84 MCP tools, 15 clients.

bkmr

Knowledge Management for Humans and Agents

gemcode

Local-first coding agent on Google Gemini + ADK

anamne

Brain-inspired local-first memory layer for Claude, Cursor, and any MCP-compatible AI tool. Three memory layers (episodic / scratchpad / working), ACT-R activation decay, 21 MCP tools. Local SQLite + ChromaDB, no SaaS.

velesdb

VelesDB is a local‑first AI data engine written in Rust that unifies vectors, full‑text and graph in a single file with a familiar SQL‑like language. Instead of sending every RAG or semantic search query to a remote cluster, VelesDB runs directly on your server, laptop, browser, mobile or edge device — no cloud dependency, no external services, ..

compair-core

Compair Core is the open-source backbone of the Compair collaboration platform. It ships the shared FastAPI app, data models, email helpers, and background utilities needed to run Compair locally or embed it in your own projects, with clean hooks for premium features layered on in the hosted cloud offering.

caliper-ai

Per-PR cost for your AI coding tools. Reads local Codex, Claude Code, Cursor, and Aider logs. Offline. No login. No SaaS. No telemetry.

superlocalmemory

World's first local-only AI memory to break 74% retrieval and 60% zero-LLM on LoCoMo. No cloud, no APIs, no data leaves your machine. Additionally, mode C (LLM/Cloud) - 87.7% LoCoMo. Research-backed. arXiv: 2603.14588

autorunne

Repo-local project memory and handoff layer for direct AI coding agents, with validation evidence and sandbox-tolerant Autorunne state write-backs.

htmlgraph

Local-first observability and coordination platform for AI-assisted development

omega-memory

Cross-model memory for AI agents. Local-first, works with Claude, GPT, Gemini, Cursor, Claw Code, and any MCP client

llama-index-vector-stores-velesdb

VelesDB is a local‑first AI data engine written in Rust that unifies vectors, full‑text and graph in a single file with a familiar SQL‑like language. Instead of sending every RAG or semantic search query to a remote cluster, VelesDB runs directly on your server, laptop, browser, mobile or edge device — no cloud dependency, no external services, ..

elastik

curl is all you need. Bytes over HTTP, packets over CoAP, one path between intelligences. Ships with CurlBench, the first LLM tool-use benchmark graded by HTTP status codes

cognithor

Cognithor · Agent OS: Local-first autonomous agent operating system. 19 LLM providers, 18 channels, 145 MCP tools, 6-tier memory, Agent Packs marketplace, zero telemetry. Python 3.12+, Apache 2.0.

stackunderflow

StackUnderflow: the local observability for your coding agents. Search, replay, and analyse every session, all offline. Starts with Claude Code.

langchain-velesdb

VelesDB is a local‑first AI data engine written in Rust that unifies vectors, full‑text and graph in a single file with a familiar SQL‑like language. Instead of sending every RAG or semantic search query to a remote cluster, VelesDB runs directly on your server, laptop, browser, mobile or edge device — no cloud dependency, no external services, ..

vectlite

Embedded vector store for local-first AI applications.

taprun

Local-first browser automation MCP for Claude Code / Cursor. taprun compiles once with AI, your authenticated Chrome replays forever at zero LLM tokens. Credentials never leave your machine.

cccc-pair

Coordinate your coding agents like a group chat — read receipts, delivery tracking, and remote ops from your phone. One pip install, zero infrastructure. A production‑minded orchestrator for 24/7 workflow

memgentic-native

Universal AI memory layer — your second brain across every AI tool. Local-first, open source, MCP-compatible.