feed · watching · stats · help with your tokens
KKKKhazix/agent-autoloop
Agent Autoloop lets a command-line agent pick up work in the background and follow it through, asking for a decision only when something is uncertain.
It addresses command-line agents that only work while someone watches the terminal, so the job stops once the window is closed or the person walks away. A kernel of about 1650 lines of code with zero runtime dependencies coordinates the work.
Similar
- vercel-labs/agent-desktop
agent-desktop gives AI agents computer use on macOS, Windows and Linux by reading the OS accessibility tree instead of screenshots.
- aaitor/skill-video-studio
A local video-production workspace driven by a Claude Code skill that turns a product walkthrough into a polished demo video.
- jyotipmahes/ML-Interview-Prep
A repository containing Machine Learning Engineer interview preparation materials and from-scratch algorithm implementations.
- postmanlabs/postman-app-support
The public support repository where Postman users file feature requests, integration ideas and bug reports.
- JustVugg/colibri
colibri is a C inference engine that runs very large open mixture-of-experts models on the machine you already have.
- deepseek-ai/dsh-superagent-prototype
dsh-superagent is a research prototype for a proactive coding agent that works on a single GitHub repository using DeepSeek Harness.
- anthropics/FlashProve
FlashProve is a research prototype for zero-knowledge proofs of AI inference that allows providers to prove model identity without revealing weights.
- Muvon/octomind
Octomind is an open-source coding agent harness that gives an LLM real access to a repository through MCP tools.
sources
Claims below include README evidence. A matching quote or worker agreement does not establish that a claim is correct. Category majority among 2 workers. Version 1.
Agent Autoloop lets a command-line agent pick up work in the background and follow it through, asking for a decision only when something is uncertain.
让你的命令行 Agent 在后台自己接活、自己跟进到底
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
It addresses command-line agents that only work while someone watches the terminal, so the job stops once the window is closed or the person walks away.
平时我们用 Claude Code、Codex 这类命令行 Agent，大多是“人守着终端用”
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
A kernel of about 1650 lines of code with zero runtime dependencies coordinates the work.
内核约 1650 行代码、零运行时依赖。
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Runs on a laptop without a cloud server or open public ports.
装在你的 Mac 或 Linux 上，不用买云服务器，也不用开放公网端口。
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Connects a new agent with one line of config, without an SDK or source changes.
**一行配置接新 Agent。** 不需要 SDK，也不需要改 Agent 源码
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Pushes decision cards to chat tools or desktop notifications.
需要你拍板时把卡片推到飞书、钉钉、企微、Slack、Discord 或桌面通知
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
While waiting for a decision the process exits, so no agent runs and no tokens are spent.
等你拍板时进程直接退出，等待状态存在数据库里，Agent 不用一直挂着。
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Rules take effect seconds after saving and can be checked with a command.
保存后几秒内自动生效，可以运行 `agent-autoloop check` 检查。
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Bundled recipe templates include rules, a task skill and setup notes.
每套模板都带规则、任务 Skill 和启用说明。
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
A delivery pipeline follows a GitHub PR tagged deliver through review, CI, merge and deployment.
给 GitHub PR 打上 `deliver` 标签后，它会自己跟进到上线
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Cards can also be viewed and answered from the CLI and MCP.
命令行和 [MCP](skills/agent-autoloop/SKILL.md#mcp-与密钥撤销) 也能查看和回答卡片。
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Turns cron schedules, new GitHub issues and business-system errors into standard events.
早上 9 点的定时任务、GitHub 上新提的 Issue、业务系统报的错
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Runs on one computer with one command
一台电脑、一条命令就能跑
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Connects a new agent with one line of config
一行配置接新 Agent
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Sends decision cards to Feishu, DingTalk, WeCom, Slack, Discord or desktop notifications
需要你拍板时把卡片推到飞书、钉钉、企微、Slack、Discord 或桌面通知
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
While it waits for your answer it uses no background process and no tokens
等你回复期间，它不占后台进程，也不消耗 Token
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: Needs Node 24, Git and curl installed before running.
运行前需要装好 Node 24、Git、curl
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: The four webhook templates were only validated against local mock services.
四种 Webhook 模板只在本地模拟服务上验证过。
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: Text channels need hub.publicUrl set to an address reachable from a phone or computer.
要把 `hub.publicUrl` 配成手机或电脑能打开的地址。
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
Limitation: macOS and Linux only; needs Node 24, Git and curl
支持 macOS 和 Linux。运行前需要装好 Node 24、Git、curl
Quote matched the README or GitHub description at extraction. Same claim and quote from 1 worker(s). Interpretation not reviewed by a maintainer.
developer-tools · ai-ml