AI middleware
-
aethercore
Provide a lightweight agent kernel designed for efficient, fast startup in distributed AI applications with minimal resource use.
AI Go -
agent-egress-bench
Provide an open test corpus to assess and improve AI agent egress security through validated cases and automated workflows.
-
aigc-cli
AIGC CLI - 统一 AI 图片、视频、Midjourney 生成与检测命令行工具。支持 OpenAI、OpenRouter 及任意 OpenAI 兼容中转。内置 MCP Server、AIGC 检测引擎。
-
alertint-agent
Read-only Alertmanager intelligence agent — correlates alerts, triages incidents with LLMs, and exposes findings via MCP to AI coding agents
-
ariadne
Local-first multilingual MCP memory for Codex and Claude Code: Qdrant + Ollama bge-m3 hybrid search, curation tools, auto-recall/capture, import/export, backups, and a cross-platform tray. No Docker or cloud.
-
aurora
Open-source AI gateway — route LLM traffic across OpenAI, Anthropic, Gemini, Groq & 30+ providers via a single API. Self-hosted, no vendor lock-in.
-
auxly-memory-cli
A local-first, file-based memory CLI for AI tools. Own your context in plain Markdown — auditable, versionable, and provider-agnostic.
-
claude-proxy-pro
Route Claude Code requests to any AI provider using a native desktop application.
-
clovapi
内置本地代理，轻松管理 Agent API — 开源 CLI 与桌面端，保存 profile、switch 到 Claude Code / Codex / OpenCode 等 Agent。
-
confluence2md
🛠️ Convert Confluence `.doc` exports to clean Markdown effortlessly with this CLI tool, streamlining your documentation process.
-
core-agent
Reusable Go base agent on the Google ADK — multi-provider (Gemini, Vertex, Claude on Anthropic + Vertex), MCP-native, with AGENTS.md, skills, and permissions wired up.
AI Go -
demarkus
De-centralized - markdown - for us: an information web adjacent markdown protocol for knowledge sharing between humans and agents, humans to humans, agents to humans, and agents to agents
-
director
Coordination ledger for coding agents: what was decided, what's still open, where you parked — durable across sessions, repos, and weeks. Memory tools store what the agent knows; Director tracks the state of the work. Run both: they don't overlap. Claude Code + OpenAI Codex.
-
documind-ai
DocuMind AI is a full-stack local AI document workflow platform built with Next.js, Go, Python FastAPI, PostgreSQL + pgvector, NATS, MinIO, and Ollama. It supports document upload, async processing, AI summaries, RAG chat, semantic search, workspaces, roles, and Docker-based deployment without paid AI APIs.
AI Go -
dryrun
PostgreSQL schema intelligence MCP server with offline linting, migration safety, query validation for AI coding assistants.
-
evolve-loop
Self-improving AI pipeline that autonomously discovers, builds, audits, and ships improvements to any codebase. 4 agents, 5 phases, eval gating, continuous learning. Plugin for Claude Code & Gemini CLI.
-
fak
fak — the Fused Agent Kernel: one Go binary for AI agent loops. Wrap Claude Code/Codex/Cursor, keep long sessions cache-efficient, route work per call, run local GGUF models, and adjudicate tool calls.
-
flynn
A secure, self-improving agent operating system in a single Go binary. Bring any model, manage local models, point it at a goal, and grant it real authority: every action is sandboxed, governed, and sealed into a verifiable, tamper-evident record an independent party can check. Runs interactive or 24/7, or embed it in your own system.
-
gaia
Gaia is a command-line interface (CLI) tool for interacting with language models via a local API. It features a beautiful terminal UI, robust configuration management, and multiple interaction modes (default, describe, code, shell) for versatile assistance with programming, system administration, and more.
-
ha-nova
AI-powered Home Assistant control. Lean relay + LLM skills for Claude Code, Codex, and OpenCode.
-
hopbox
A compute-box substrate: one daemon turns a host into a fleet of isolated Firecracker microVMs reached over plain SSH — for humans and AIs. Your SSH key is your identity; the username is the box spec.
AI Go -
howl
⚡ Blazing-fast Go statusline HUD for Claude Code — 13 metrics, adaptive layouts, zero dependencies
-
idea
Worktree-aware idea capture and backlog tracker — markdown-first, CLI-native
-
Incident-Investigator
Stateful MCP server for incident investigation: dynamic evidence planning, competing hypotheses, confidence scoring, timelines, blast radius, and postmortems. Works with Cursor, Claude Code, Codex, and any MCP client.
-
linear-cli
Manage Linear issues directly from your terminal with this lightweight, dependency-free binary featuring over 35 commands.
-
M31A
The terminal-native AI coding agent that ships, not just suggests. Six-phase workflow, git rollback chain, zero telemetry. One static binary.
-
mcp-yandex-tracker
Manage and automate tasks in Yandex Tracker using a robust MCP integration for efficient issue tracking and project control.
AI Go -
mcpfs
Mount MCP servers as filesystems to access reads via standard file commands and perform writes through a CLI tool.
-
mitos
Millisecond microVM sandbox forking for AI agents on Kubernetes. Firecracker VMs that restore from memory snapshots in milliseconds, fork a running VM into N copies, and persist durable, versioned workspaces. Self-hostable, declarative CRDs.
AI Go -
mnemos
Self-hosted memory + evidence layer for AI agents (Claude Code, Codex, Hermes, ...) — embeddable Go library, MCP / HTTP / CLI, evidence-backed claims, bitemporal recall, axi-go execution kernel with JSONL audit + token budgets, cosign-signed releases with SLSA L3 provenance. No vendor cloud, no per-call billing.
-
mora
Local-first, zero-egress memory for MCP agents: Gmail, Calendar, iMessage, Apple Calendar → entity graph + ranked daily brief + hybrid retrieval
-
nano-banana-cli
🖼️ Generate and edit images with ease using the Nanobanana CLI, powered by AI and Google's Gemini models for advanced image manipulation.
-
neuralvault
Self-hosted AI memory and knowledge platform for contextual retrieval. Hybrid semantic + keyword search powered by Qdrant and PostgreSQL, with multi-workspace support and BYOK multi-provider flexibility.
-
nothumansearch
Search engine for AI agents — ranks sites by agentic readiness (llms.txt, OpenAPI, MCP, ai-plugin). MCP server, REST API, full-text search. 8,000+ indexed sites.
-
nucleus
AI-first Go microservice kernel for contract-driven service generation, evolution, and verification.
-
octo-agent
A functionality-first AI agent, distributed as a single Go binary.
-
ollama-mesh
The control plane for self-hosted AI inference. Warm-state GPU routing, multi-runtime orchestration across Ollama, vLLM, llama.cpp, and TGI. Single Go binary. Apache-2.0.
-
operatorlm
Route LLM API requests through a local proxy with multi-account support, failover, and secure key storage.
-
portwing
Security-first remote Docker agent — authenticated Docker API proxy with outbound edge mode, Ed25519 per-request auth, and a cosign-signed, scratch-based supply chain. Drydock-native + generic REST.
AI Go -
Repo2AI
Convert code repositories into structured, size-limited context packs for AI tools like ChatGPT, Claude, and Cursor.
-
sequify-server
AI assistant SDK for any website. Users control your app with natural language — no screenshots, no DOM parsing, no guessing. Websites register capabilities, AI calls them directly.为任意网站接入 AI 助手的 SDK。用户用自然语言操控你的应用 — 不截图、不解析 DOM、不靠猜。网站主动注册能力，AI 直接调用。
-
symaira-vault
🔐 The password manager for terminal users and AI agents. Age-encrypted, keyring-cached, MCP-ready. Zero telemetry.
-
taskrail
Deterministic execution harness for AI agents — goals in, structured tasks out, verified results back.
-
tela
Open-source, self-hostable markdown team wiki with a built-in MCP server — so AI agents read, write, and search your docs as first-class teammates. Live multiplayer, semantic + full-text search. Go + PostgreSQL + React/Milkdown.
-
tokenctl
Hierarchical org→team→dev token-budget proxy for LLM traffic: meter, soft-throttle, hard-deny, and in-flight preemption across providers on one shared wallet.
-
trip2g
Open-source MCP knowledge mesh - self-host your second brain, expose it to AI agents, federate with peers.
-
Unpinched
Detect PinchTab deployments and browser bridge artifacts with a fast, point-in-time scanner for security and forensic analysis.
-
versions-skills
AI-native version number toolkit in Go — Skills + MCP + SDK + CLI. Parse, compare, sort, group, check constraints & range-query version numbers. Works with Claude Code, Codex, Cursor, Windsurf, Cline, VS Code Copilot.
-
vibescript
An embeddable Ruby-like language for building safe, AI-friendly applications in Go
AI Go -
vswarm
Multi-tenant T3 Code behind a single URL, gated by Cloudflare Access — each user gets an isolated browser workspace, no pairing, no Tailscale.
-
wick
Build internal tools and AI agents in Go — or just download and run Claude / Codex / Gemini as a Slack + Telegram + web agent host. No copy-pasting. You own the code.
AI Go -
workcell
Bounded local runtime and policy boundary for coding agents
AI Go -
wt
Git worktree CLI for parallel-edit workflows — sibling layout and shell-integrated navigation
-
you-agent-factory
Agent factory workflow engine for agents
AI Go -
KindleVibe