#llm

Open-source coding CLI for repository-native, multi-model, verifiable engineering workflows.

Coding should feel like thinking. A terminal coding-agent runner: interview → reviewed plan → verified execution, tmux workers, multi-provider fallback, GitHub/GitLab tools, and verification receipts.

MCP server in Go to search, download and read books, papers, comics and more from Library Genesis — plus keyless open-access discovery (arXiv, Crossref, OpenLibrary, Gutenberg, dblp, PubMed, ERIC) and article downloads via Unpaywall, Europe PMC, bioRxiv, CORE and OAPEN. Four focused tools, one static binary, no account required.

An open-source TypeScript runtime for building and running agents.

PostgreSQL compiler. One source, many faithful forms.

MCP server providing read-only Helm chart tools for LLM agents

Your work, already underway. A personal AI assistant for your job: mail, chats and tickets become tasks, the agents you already use (Claude Code, Codex, Gemini) do the work, and nothing goes out until you approve. Open source, runs on your machine.

Open-source pipeline for rewriting AI-generated text into natural human prose. Need it in your app? Try the Lynote Humanizer API → lynote.ai/ai-humanizer-api

A Model Context Protocol (MCP) server implementation that connects Large Language Models (LLMs) to GIS operations using GIS libraries, enabling AI assistants to perform geospatial operations and transformations.

Local-first CLI (and desktop) for token and cost analytics across AI coding agents

AgentWorks: AI agents that own the work. Goals that chase a metric, and Crews your team asks from Claude, ChatGPT or Cursor. Open source.

The governance layer LLM proxies promised as a TypeScript library you can read. Every call runs through one governed pipeline: rate limit → spend caps (per-user + global circuit breaker) → cache → schema-validated provider failover → usage log → LLM-judge No sidecar proxy to deploy.

Shared, persistent memory for AI assistants, built on the Zettelkasten method

The next generation of linting: an LLM as a judge for code-quality checks deterministic linters can't express. Built on oneharness.

Translate EPUB books using Large Language Models while preserving the original text. The translated content is displayed side-by-side with the original, creating bilingual books perfect for language learning and cross-reference reading.

Contains parallel-cli and the parallel-web-tools Python package. Built on the parallel-web SDK: https://github.com/parallel-web/parallel-sdk-python

PubNub MCP Model Context Protocol Server for use in Cursor, Windsurf, Claude Desktop, Claude Code and OpenAI Codex and more!

AI sovereignty for the world's most trusted runtime.

A lightweight library for interacting with Google's Gemini API from C#/.NET including Microsoft.Extensions.AI support.

jq-like stream processor for semantic matching and extraction

Chat with your database or your datalake (SQL, CSV, parquet). PandasAI makes data analysis conversational using LLMs and RAG.

Merge heterogeneous chat/text sources into a single LLM training format (JSONL)

After the spec: ship and run. A delivery framework for agentic systems — floor first, evolution on evidence, governance from day zero, handover with proof.

Superagent protects your AI applications against prompt injections, data leaks, and harmful outputs. Embed safety directly into your app and prove compliance to your customers.

🥤 RAGLite is a Python toolkit for Retrieval-Augmented Generation (RAG) with DuckDB or PostgreSQL

PHP library for interacting with AI platform provider.

The taigrr fork of Charm™️'s crush

A feature-full starting point for production agent applications on Tangle.

A POSIX shell with AI in the grammar: type ?? to drive Claude Code, Cursor CLI, Gemini CLI, or Ollama over ACP. Local-first, no login, no telemetry.

Source generator to convert C# interfaces to JSON Schema for AI tool calling