Havuz

The lossless, headless coding agent harness. One 7 MB static binary, four tools, plain stdin/stdout, no silent cuts.

Run language-model agents in Minecraft with Mineflayer tools, persistent sessions and shell sandboxes.

A coding agent that rewrites its own harness (prompts, tools, compiled core) while it works, so the harness grows with the model. A Root it

Free open-source toolkit to understand, audit and rescue vibe-coded or legacy codebases: source-linked maps, team handoff, human-approved fi

Make AI agents aim higher without losing control: a Claude Code skill with two bars per task, a cheap gate check before expensive work, and

A recursive self-improvement (RSI) workflow that evolves both target agents (e.g., Pi and DeepSeek Harness) and the RSI system itself.

An AI agent plays Baldur's Gate 3 on Linux through a Lua mod and an MCP server, seeing only what a player sees. Includes a model benchmark a

The harness that builds harnesses: tell it about your work and it builds a custom AI agent with its own tools, views, skills and UI. Runs in

Let an AI agent use your PS Vita.

The last agentic harness you'll ever need. Bring any model; Pennant does the rest on your Mac and iPhone, and asks before anything is sent,

Native DeepSeek Harness (DSH) plugin integrating TypeSafe Jev or Decision api like luna as a System One decision layer for agent selection,

DSH 插件：把 WorkBuddy 桌面 App 里的模型接入 DeepSeek Harness，零配置直接用。（原生嵌入"插件-插件配置页"）

A minimal yet ready-to-use AI agent harness, built on best practices for frontier models.

mu, coding agent kararlarının bir kısmını küçük bir judge modeline devrederek pi tabanlı CLI ve desktop agent sunar.

Neden yükseliyor · 14 günlük yaşına rağmen 386 yıldız, son 28 günde 731 commit ve 11 farklı yazar raporlanıyor; bu kısa sürede dikkat ve geliştirme hareketliliğine işaret ediyor. Ancak bu veriler kalıcı benimseme veya üretim olgunluğunu tek başına kanıtlamıyor.

Claude Code-style auto-memory plugin for DeepSeek Harness (dsh): typed memory files + MEMORY.md index auto-injected into the system prompt.

A custom coding harness for TypeSafe AI's Jev: an LLM proposes, Jev answers narrow questions, code decides, every step leaves a receipt.

⚡ Millisecond System-1 judgement for every tool call in DeepSeek Harness — Jev-powered risk classification & evidence-gated auto-approval. F

Jev karar motorunu Agent Harness ortamlarına eklenti olarak bağlayıp araç, Skill veya görev sorumlusu seçimi için yapılandırılmış değerlendirme sunuyor.

Neden yükseliyor · Repo yalnızca 8 günlükken 215 yıldız almış olması kısa sürede dikkat çektiğini gösteriyor; ancak bunun benimsemeye mi yoksa geçici ilgiyi mi yansıttığını anlamak için veri yetersiz. Son 28 gündeki 7 commit’in tek yazardan gelmesi de topluluk katkısı konusunda henüz güçlü kanıt sunmuyor.

Harness-Zero: Harness Distillation via Agent-as-Harness

Coding-agent tools on TypeSafe's Jev, measured in public. review-router: caught 13 of 13 CVE fixes a filename rule sent to a quick review. j

Jev-powered decision layer for DeepSeek Harness

Local-first agent harness (Windows + Ollama) for 2–9B models. Same 2B model: 0.017 → 0.821 across four agent harnesses — 48×. 288-cell bench

Your AI Video Harness. Runs locally, with a visual editor so you can jump in anytime.

The control plane for your coding agents, however you run them. One checkout of rules, skills, subagents, hooks and stances, projected into

Give your coding agent a delivery process you can trust. Spec → Plan → Build → Acceptance → Ship.

The open archive of how AI agents actually work. A versioned, verifiable archive of system prompts, developer instructions, tool schemas an

Redpen makes coding agents prove they are actually done.

DSH 插件：基于知乎开放平台官方 API 创建的站内搜索、全网索引搜索与直答三个工具。搜索不幻觉，引用有出处。（原生嵌入“设置-插件-插件配置”）

Benchmark coding agents and agent harnesses on 226 deterministic file-editing tasks with byte-exact verification and a public leaderboard.

Execution posture harness for Claude Code. Your coding agent stops one step early; Castra is the step.

Nausicaa treats an Agent run as a dynamic topology of Lanes rather than one fixed linear loop. The model can decide when to observe, fan out

Entente — provenance engineering for coding agents: handoffs become confirmable, verifiable, retryable contracts; agents form a network with

Turn one ZCode conversation into a restartable multi-agent workflow with persistent memory, isolated coding agents, real-result checks, and

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。

Self-improving agent harness for ChatGPT and Codex on Mac. #1 on AssistantBench. On-disk memory, privacy zones, verified completion.

A harness for hosted Grok Bot crews. Its core rule: it is not allowed to guess.

Gray is an open-source (MIT) AI agent harness shipped as one static Rust binary. Bash-only tools, resumable JSONL sessions, a self-compactin

Spec Driven — orchestrator-first Agent Skills for spec-driven development. The orchestrator skill turns a PRD or idea into approved SPECs, t

Five layers of an agent harness, each with the failure it prevents. Runnable with no API key.

Minimal surface, complete architecture: a small, local-first coding-agent harness that re-derives the invariants of DeepSeek Harness with fe

Independent evidence-gated second brain for understanding and porting DeepSeek Harness architecture.

The operational superset of Pi Coding Agent — everything Pi, plus observability, governance, recovery, evaluation and multi-agent orchestrat

:dog: Beagle is designed to be the unified platform for the broader Recursive Self-Improvement (RSI) ecosystem. Think of it as Hugging Face

The best AI harness you can run yourself. One file: web UI, tools, memory and orchestration built in. Run a team of agents across providers

A coding agent harness built specifically for Motif-3

Scikit-learn for Recursive Self-improvement (RSI). A modular framework for recursive harness evolution.

Durable single-Agent Harness for TypeScript: recoverable Threads, context continuity, explicit side effects, and a native TUI.

Your AI coding harness, ported to every client you use. Capture Claude Code (or Codex) rules, hooks, skills, subagents, commands and MCP ser

A terminal where one Claude Code session supervises the AI agents in your other tabs: which is stuck, which is done, which is still running.

Safe to explore. Free to build. Koder by aixplain is an AI coding harness for desktop and terminal, with permission checks, reviewable chang