A curated directory of more than 2,500 production-ready APIs for building autonomous AI systems, organized into Agents, AI Models, and MCP Servers with links to providers and docs.cporter202
Llm-wiki-compiler compiles raw sources into a citation-traceable, interlinked wiki that humans and agents can browse, query, and reuse via CLI, MCP, or SDK.atomicstrata
A local-first token usage and cost tracker for dozens of AI coding tools, with native desktop apps, widgets, quota monitoring and optional cross-device sync.xiufengsun
Efficient mixture-of-experts foundation model weights from Tencent Hunyuan for real-world embodied agents, balancing physical understanding, reasoning performance, and inference efficiency.tencent-hunyuan
A self-hosted AI agent platform combining multi-framework agent deployment, model management, an app store, memory, and distributed compute clustering across your own hardware.jaylfc
ApexYard is a forkable ops repo that governs AI coding agents with roles, shell hooks, slash-command skills and merge gates, so agent-built changes follow a ticket-first SDLC and reach production safely.me2resh
An open-source job-search skill that runs inside your AI coding CLI to score job posts, tailor CVs, draft cover letters, and track applications.santifer
career-ops is a job-search skill for AI coding CLIs. It scans career pages, scores roles against your CV, tailors ATS-friendly resumes and tracks applications, while leaving every submission decision to you.career-ops-hq
An open-source AI penetration-testing and bug-bounty agent that autonomously investigates web applications and verifies exploits with machine oracles, usable via MCP, CLI or CI.0xSteph
Generates compact codebase context and wiki knowledge for AI coding assistants, and exposes code-structure lookups as an MCP server to cut token waste.Houseofmvps
look is a keyboard-first, local-first launcher for macOS, Windows and Linux. It opens apps, files, folders and clipboard history instantly, with no account, telemetry or background daemons.kunkka19xx
A developer package from Web Kits offering declarative audio synthesis for the web, complete with a CLI for authoring sound patches; installed from npm and imported into front-end projects.raphaelsalaja
Batteries-included framework for building web applications in Rust, providing routing, components, and server-side rendering for complete full-stack development aimed at web developers.tokio-rs
A unified SQL interface built for AI agents, letting developers query APIs, files, and live sources together through MCP without building separate connectors.withcoral
Fast LLM inference server for consumer-grade GPUs that uses speculative decoding and custom kernels to raise throughput while lowering memory use.luce-org
A local-first command-line tool that syncs, searches and organizes X bookmarks, exports a Markdown knowledge base, and exposes local context to Claude Code, Codex or any shell-capable agent.afar1
A shared memory and skill layer for coding agents. It captures sessions, indexes traces for recall, and mines repeated patterns into reusable SKILL.md files across Claude Code, Codex, Cursor and others.activeloopai
Collaboration layer letting people and multiple AI coding agents work in shared rooms, with identities, inboxes, shared memory, and cross-room context for coordinated development.alookai
termcn is a library of ready-to-use, theme-aware terminal UI components for React, built on Ink and OpenTUI and compatible with the shadcn/ui registry format.shadcn-labs
Native .NET inference engine for GGUF text and multimodal LLMs, providing a console app, web chatbot interface, and Ollama/OpenAI-compatible APIs with GPU support across Windows, macOS, and Linux.zhongkaifu
⚡ DocsAgent - give your AI agents instant, private access to your personal knowledge base (Zotero, Obsidian, Apple Notes supported now, local docs on the road). Native C++ search core: BM25 + passage ranking in ~15 ms. MCP server for Claude, Cursor, Cline & any MCP client.docsagent
A unified toolkit for multimodal model inference, evaluation, and post-training, supporting over a dozen models and a hundred benchmarks. Serves researchers extending evaluations and training tasks to the cloud.aifrontierlab
An open-source menu-bar companion for heavy coding-agent users, monitoring multiple Claude Code, Codex, and OpenCode sessions with approval alerts and one-click terminal switching.octane0411
Cross-vendor messaging for CLI coding agents: Claude Code, Codex, Gemini, and others exchange messages through a shared local SQLite store, letting separate agent sessions collaborate without copy-paste or a broker.fujibee
A local-first workbench that runs Claude Code, Codex, Cursor and other coding agents in parallel isolated git worktrees, with built-in review, testing and pull-request shipping.dohooo
Agent skill turning coding assistants like Claude and Codex into geospatial analytics agents, supporting spatial SQL queries with cost guards and map-based visual correction.dekart-xyz
Maestro is an agent-workflow skill for AI coding tools. It provides one core skill, 25 commands, domain references, persistent memory, and an audit trail for designing and improving agent workflows.sharpdeveye
An open-source terminal coding agent that works with OpenAI-compatible APIs, Gemini, Codex, Ollama, and many other providers while keeping prompts, tools, agents, MCP, and slash commands in one workflow.Twigpine
Lightweight agent runtime shell with a built-in personal assistant, offering dozens of tools, skill memory, and multi-agent collaboration for users.hkuds
OpenRig turns separate Claude Code and Codex sessions into persistent, YAML-defined agent teams. It boots, coordinates, and recovers multi-agent topologies across terminals, with shared context and owned work.mvschwarz
Two online books on harness engineering, explaining how Claude Code and Codex organize constraints, loops, permissions, and verification for reliable coding agents.wquguru
A runtime patcher layered on top of official Claude Code that unlocks hidden features, removes selected restrictions, and re-applies itself after updates.0Chencc
A coding-agent skill that packages academic search workflows, helping agents retrieve papers, parse content, and support structured reading for researchers and developers.ustc-ai4science
A macOS Dynamic Island companion that monitors Claude Code sessions, handles permission approvals, and syncs status to iPhone for remote follow-up.MioMioOS
A text-to-speech model and application suite that provides high-quality zero-shot voice cloning and voice design across more than six hundred supported languages.k2-fsa
Claurst is a Rust-built, multi-provider terminal coding agent with a rich TUI, plugins, skills, sandboxed shell access and headless mode, capable of reading codebases, editing files and running commands autonomously.Kuberwastaken
An overnight orchestrator that loops coding agents toward an objective with incremental commits, retries, worktrees, and multi-agent support.kunchenguid
CC Gateway is a reverse proxy that sits between Claude Code and the Anthropic API, normalizing device identity, environment fingerprints, and telemetry for privacy.motiful
Muxy is a lightweight, memory-efficient terminal multiplexer with split panes, background and remote sessions, git worktrees, and integrations for AI coding agents.muxy-app
A local-first AI desktop client with built-in llama.cpp inference, multi-model management, MCP support, and an OpenAI-compatible API for offline use.AtomicBot-ai
A reconstructed Claude Code release recovered from published npm source maps, bundling a runnable CLI, TypeScript source, and official plugins for study or direct use as a coding assistant.LING71671
A structured three-role development workflow pack for coding assistants, adding architect, builder, and reviewer discipline with token-saving rules.russelleNVy
Scarf is a native macOS and iOS app for the Hermes AI agent. It provides dashboards, chat, session history, memory and skill editing, cron and MCP management over local or SSH connections.awizemann
A Claude Code plugin that interviews you, designs the stack and architecture, and writes self-contained blueprints with EARS acceptance criteria and verifiable build steps.Hainrixz