Library
85,762 resources from 11 public sources. 53 checked by hand; the rest are unreviewed.
Results
Looks like a goal we have a starting path for: I want to build an AI agent
-
dsh-auto-memory
Agents & frameworksTrustworthy memory for DeepSeek Harness — 4-state trust machine, evidence chains, audited injections. Zero servers, just Markdown.
git clone https://github.com/AskTheWay/dsh-auto-memory.git -
localharness
Agents & frameworksAn open-source, coding agent harness designed specifically for local LLMs. Designed as everything a user needs to jumpstart their own harness (detailed event steams, logs, and performance data) or quickly replace closed source one (featuring out of the box plugins for web search, image gen, memory, auto research and more in dev).
git clone https://github.com/ahwurm/localharness.git -
Agent-App
SkillsAgent App is application that AI agent can build, evolve, and operate. A collaboration space for both humans and agents.
git clone https://github.com/CraftOS-dev/Agent-App.git -
ruflo
MCP servers🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
git clone https://github.com/ruvnet/ruflo.git -
open-astravia
Skills开源 · 本地优先的 AI 智能体:编程与实干兼备。 自带模型密钥,内置 MCP、技能、插件、工作流、知识库以及数据库。
git clone https://github.com/maomaochong-ai/open-astravia.git -
Orkas
Agents & frameworksOrkas is an open-source (MIT), local-first AI team for your desktop: a Commander turns your goal into to-dos and coordinates 16 built-in specialist agents plus Claude Code, Codex, OpenCode, OpenClaw and Hermes, with projects, scheduled automations and connectors. Free app for macOS & Windows; built-in models or your own keys.
git clone https://github.com/Orkas-AI/Orkas.git -
BossConsole
MCP serversOpen-source, multi-platform harness for AI agents - a native, multi-threaded operator's console (JVM, not Electron) to run Claude Code, Codex, Gemini or OpenCode with a real browser, terminal, editor, secrets & 200+ MCP tools. Built for enterprises, science & research.
git clone https://github.com/risa-labs-inc/BossConsole.git -
alloy
Agents & frameworksModel-agnostic agent harness for Elixir
git clone https://github.com/alloy-ex/alloy.git -
speq-skill
SkillsA light-weight and straightforward workflow for spec-driven development with Claude Code and Codex. Written in Rust 🦀
git clone https://github.com/marconae/speq-skill.git -
harness-sdk
Agents & frameworksBuild an agent harness and control it end-to-end. Open-source SDK for production AI agents in Python & TypeScript - any model, any cloud.
git clone https://github.com/strands-agents/harness-sdk.git -
garcon
Agents & frameworksSelf-hosted browser workspace to run coding agents in parallel, steer work as it runs, review diffs, and ship.
git clone https://github.com/cfal/garcon.git -
OpenAdapt
Tools & appsFor EHRs, payer portals, and any app your team clicks through. A person or an agent shows the task; OpenAdapt turns it into a local program that repeats the steps with no model calls, checks each save, and stops when something doesn't match.
git clone https://github.com/OpenAdaptAI/OpenAdapt.git -
waku-agent
Agents & frameworksWaku Waku! Waku Agent is a local-first AI agent harness you actually own, including loop, memory, eval, all in code built to stay legible as it grows.
git clone https://github.com/ShenSeanChen/waku-agent.git -
mi
Tools & appsself-extending sandboxed agentic harness. a loop, two tools and an llm
git clone https://github.com/av/mi.git -
Carbon-Monitor
SkillsAgentic PR reviews for the IBM Carbon Design System, run by IBM Bob or Claude from a codified review prompt and grounded by Carbon MCP. Built on a reusable agent harness for IBM Cloud Code Engine: scheduled, parallel runs, a per-run spend cap, a web console, Elasticsearch telemetry and a Slack digest. A skill adapts it to your repo, job or agent.
git clone https://github.com/carbon-design-system/Carbon-Monitor.git -
clawboo
Agents & frameworksAn open-source studio for teams of AI agents, from marketing squads to dev teams to research crews. Deploy a team, then watch them delegate and collaborate live. Self-hosted, one command.
git clone https://github.com/clawboo/clawboo.git -
rel-ai-chatgpt-web-harness
MCP serversLocal agency/runtime harness for ChatGPT Web: repositories, tasks, commands, validation, Git, skills/memory, observability, and opt-in computer control.
git clone https://github.com/Kyne0328/rel-ai-chatgpt-web-harness.git -
pioneer
SkillsMultiplayer harness where humans and AI agents work together and build on shared context.
git clone https://github.com/pioneerdotai/pioneer.git -
SharpClawCode
Tools & appsC# coding-agent harness (.NET 10, Microsoft Agent Framework) — sessions, tools, permissions, providers, MCP, plugins
git clone https://github.com/clawdotnet/SharpClawCode.git -
skill-evolution
SkillsImprove Claude Code and OpenAI Codex skills from real feedback, with evidence, regression checks, and human review.
git clone https://github.com/rwang23/skill-evolution.git -
forge
SkillsGive your coding agent a delivery process you can trust. Spec → Plan → Build → Acceptance → Ship.
git clone https://github.com/brightstack/forge.git -
learn-agent-the-hard-way
Docs & tutorials笨办法学 Agent · 亲手打造一个 harness——不用框架,32 个练习亲手写出一个 agent harness 的每一层。系列第一本,全书已完结,三本各自独立不分先后。
git clone https://github.com/Leihb/learn-agent-the-hard-way.git -
pi
Agents & frameworksA pure-Go port of pi (earendil-works/pi) — the pi-ai unified multi-provider LLM API, pi-agent-core agent loop, and pi-coding-agent, native in Go with no Node.js runtime. Anthropic/OpenAI/Google, 999-model catalog.
git clone https://github.com/sky-valley/pi.git -
openbench
Datasets & benchmarksSame model, different wrapper: a from-scratch benchmark comparing coding-agent harnesses (codex, pi, opencode, cursor, devin) and open models on correctness, speed, and token cost
git clone https://github.com/minghinmatthewlam/openbench.git
Indexed entries aren't reviewed or tested by InnerVault. Details come straight from each source. Anything that can reach your accounts, files or browser can do damage, so read the source before you install it.