#88th of 54 in Agent and MCP starters
agent-autoloop
Background scheduler that runs CLI agents on events and asks you only when unsure
- Stars
- 4.5k
- License
- MIT
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- TypeScript
Checked on GitHub.
Overview
Agent Autoloop is a zero-runtime-dependency Node hub that dispatches events (cron, GitHub/GitLab webhooks, HTTP) to command-line agents such as Claude Code, Codex, ACP agents or any command, per rules in rules.json. When an agent needs a decision, the process exits and a card goes to Feishu, DingTalk, WeCom, Slack, Discord or desktop notification. After you answer, the original session resumes. Ships recipes for daily briefs, issue handling, alert triage and a PR delivery pipeline.
Who it is for: Developers running Claude Code or Codex unattended with human approval
Strengths
- Waiting for approval uses no agent process and no tokens; the session resumes after you answer
- Any CLI agent that takes a prompt and prints a reply connects with one config line
- Append-only event log; the `why` command explains any run; crash and simulation tests included
- Event text is passed as data files, not prompts; events cannot set cwd, agent or access
Weaknesses
- No web UI; status and answers go through CLI commands and chat cards
- README is Chinese first; English version is in a separate file
- Webhook channels and several agent commands only tested against mock servers or docs, not live services
- Text-channel approvals need hub.publicUrl set to an address your phone can reach
What it needs
- TypeScript, Claude Code, Codex, ACP agents (Gemini CLI, OpenCode), DeepSeek Harness
- Needs Node 24+, Git, A logged-in CLI agent (Claude Code, Codex, etc.), curl (macOS/Linux install)
Quick answers
Is agent-autoloop open source?
Yes. Its code is on GitHub under the MIT license.
Also in Agent and MCP starters
See all 54| Rank | Project | Score |
|---|---|---|
| 1 | deepseek-harnessPlugin-based agent harness from DeepSeek with a local Web UI | 98 out of 100 |
| 2 | composioTypeScript and Python SDKs that give agents authenticated tools across 1000+ apps | 92 out of 100 |
| 3 | conductorDurable workflow engine that orchestrates microservices and LLM agents from JSON | 91 out of 100 |
| #4 | mcp-useTypeScript framework for building MCP servers with React Views and an Inspector | 87 out of 100 |
| #5 | a2uiProtocol and renderers for agents to stream native UI instead of text | 86 out of 100 |