GanyuanRan/Autoloom

AI coding with Aegis governance built into execution: baseline-aware changes, evidence-backed delivery. Free desktop client, your choice of model. 将哲科思维融入 AI 开发执行，让变更有依据、交付有证据。用创意编织现实。

What it solves

Coding agents often plow ahead with changes without pausing to ask whether a change is even needed, whether it adds unnecessary complexity, or whether the result actually matches the requirements and evidence. Autoloom addresses this by embedding engineering governance directly into the coding runtime, so the agent questions assumptions, checks for structural bloat, and reviews delivery against verified facts—reducing the need for you to constantly remind it to investigate first.

How it works

Autoloom is a Windows desktop client built on DeepSeek's official DeepSeek Harness, with Aegis governance methods woven into the workflow. At key points—task start, before edits, after failures, and before recording results—it applies specific methods like first-principles reasoning, anti-entropy checks, counterexample review, and completion verification. It maintains task state, prerequisites, and execution evidence, and shows a collapsed governance impact row in the conversation whenever a project-changing action is about to run. It also records project goals and decisions, and lets you set test-first preferences and complexity thresholds.

Who it's for

Developers and teams using AI coding agents on Windows x64 who want more autonomous development without losing engineering judgment. It's for people who are tired of re-explaining context and re-checking results, and who want their agent to demonstrate what it actually verified. macOS and Linux versions are planned but not yet available.

Highlights

  • Integrates Aegis governance: change necessity, complexity checks, causal diagnosis, and delivery review.
  • Built on DeepSeek Harness (source commit b150a551) with the Cordis plugin mechanism retained.
  • Works with local projects and Linux x64 projects over SSH, with diffs, command logs, previews, and docked terminals.
  • Includes built-in web search via Parallel/Exa MCP services, no separate API key needed.
  • Records process evidence and allows inspection of decisions and diffs.
  • Currently in Beta for Windows x64; downloadable installer and SHA256 checksums provided.

Related

  • Project
    StanleyNull/AutoHunter

    AutoHunter is an AI‑driven, Docker‑based platform that continuously hunts for web vulnerabilities. It uses LLM‑guided agents (Collector, Worker, Reviewer) together with real tools like nmap, nuclei, and sqlmap. Humans only perform final review. The system supports multiple search engines (FOFA, Quake, Shodan, etc.), offers a Vue dashboard, and persists data in SQLite. It’s licensed Apache 2.0 and intended for authorized security testing.

  • Project
    valkor-ai/loom

    Loom is an open‑source delivery harness that works alongside AI coding agents (Claude Code, Codex, OpenCode). It adds a stateful delivery loop—scope, plan, build, test, review, repair, preview—persisted in a project‑local `.loom/` folder, enabling reliable, resumable feature development and deployment previews.

  • Project
    ningbainb/deepseek-harness-desktop

    DeepSeek Harness Desktop is a cross‑platform (Windows, macOS, Linux) open‑source desktop client that bundles the DeepSeek AI‑coding runtime with a full IDE‑like UI. It provides LLM chat, code editing, Git integration, extensible Skills/plugins, agent‑driven task automation, remote‑SSH development, and a privacy‑first design, all packaged under a BSD‑3‑Clause license.

  • Project
    theJayTea/WritingTools

    Writing Tools is a free, open‑source desktop app for Windows, Linux and macOS that lets you invoke an LLM anywhere on your computer with a hot‑key. It can proofread, rewrite, change tone, translate, summarise web pages or YouTube transcripts, and even chat with the model. You can use cloud APIs (Gemini, OpenAI, Anthropic, etc.) or run local models via Ollama, llama.cpp or Apple‑silicon MLX. The app is privacy‑focused (no telemetry, local storage of API keys) and licensed under GPL‑v3.

  • Project
    JasonLiu0826/ackem

    Ackem is a Windows‑only Electron app that turns any OpenAI‑compatible LLM into a persistent, local‑first AI companion. It offers chat, searchable memory, emotional/relationship tracking, optional desktop pet, and an extensible plugin system, while keeping all data (including API keys) on the user’s machine. The project is actively maintained, released under AGPL‑3.0, and requires no Node.js runtime for end‑users.