[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-74851":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":17,"stars30d":18,"stars90d":16,"forks30d":16,"starsTrendScore":19,"compositeScore":20,"rankGlobal":10,"rankLanguage":10,"license":21,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":22,"hasPages":22,"topics":24,"createdAt":10,"pushedAt":10,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":16,"starSnapshotCount":16,"syncStatus":28,"lastSyncTime":29,"discoverSource":30},74851,"antfarm","snarktank\u002Fantfarm","snarktank","Build your agent team in OpenClaw with one command.","https:\u002F\u002Fantfarm.cool",null,"TypeScript",2479,447,18,45,0,5,28,15,75.25,"MIT License",false,"main",[],"2026-06-12 04:01:16","# Antfarm\n\n\u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fsnarktank\u002Fantfarm\u002Fmain\u002Flanding\u002Flogo.jpeg\" alt=\"Antfarm\" width=\"80\">\n\n**Build your agent team in [OpenClaw](https:\u002F\u002Fdocs.openclaw.ai) with one command.**\n\nYou don't need to hire a dev team. You need to define one. Antfarm gives you a team of specialized AI agents — planner, developer, verifier, tester, reviewer — that work together in reliable, repeatable workflows. One install. Zero infrastructure.\n\n### Install\n\n```bash\ncurl -fsSL https:\u002F\u002Fraw.githubusercontent.com\u002Fsnarktank\u002Fantfarm\u002Fv0.5.1\u002Fscripts\u002Finstall.sh | bash\n```\n\nOr just tell your OpenClaw agent: **\"install github.com\u002Fsnarktank\u002Fantfarm\"**\n\nThat's it. Run `antfarm workflow list` to see available workflows.\n\n> **Not on npm.** Antfarm is installed from GitHub, not the npm registry. There is an unrelated `antfarm` package on npm — that's not this.\n\n> **Requires Node.js >= 22.** If `antfarm` fails with a `node:sqlite` error, make sure you're running real Node.js 22+, not Bun's node wrapper (see [#54](https:\u002F\u002Fgithub.com\u002Fsnarktank\u002Fantfarm\u002Fissues\u002F54)).\n\n---\n\n## What You Get: Agent Team Workflows\n\n### feature-dev `7 agents`\n\nDrop in a feature request. Get back a tested PR. The planner decomposes your task into stories. Each story gets implemented, verified, and tested in isolation. Failures retry automatically. Nothing ships without a code review.\n\n```\nplan → setup → implement → verify → test → PR → review\n```\n\n### security-audit `7 agents`\n\nPoint it at a repo. Get back a security fix PR with regression tests. Scans for vulnerabilities, ranks by severity, patches each one, re-audits after all fixes are applied.\n\n```\nscan → prioritize → setup → fix → verify → test → PR\n```\n\n### bug-fix `6 agents`\n\nPaste a bug report. Get back a fix with a regression test. Triager reproduces it, investigator finds root cause, fixer patches, verifier confirms. Zero babysitting.\n\n```\ntriage → investigate → setup → fix → verify → PR\n```\n\n---\n\n## Why It Works\n\n- **Deterministic workflows** — Same workflow, same steps, same order. Not \"hopefully the agent remembers to test.\"\n- **Agents verify each other** — The developer doesn't mark their own homework. A separate verifier checks every story against acceptance criteria.\n- **Fresh context, every step** — Each agent gets a clean session. No context window bloat. No hallucinated state from 50 messages ago.\n- **Retry and escalate** — Failed steps retry automatically. If retries exhaust, it escalates to you. Nothing fails silently.\n\n---\n\n## How It Works\n\n1. **Define** — Agents and steps in YAML. Each agent gets a persona, workspace, and strict acceptance criteria. No ambiguity about who does what.\n2. **Install** — One command provisions everything: agent workspaces, cron polling, subagent permissions. No Docker, no queues, no external services.\n3. **Run** — Agents poll for work independently. Claim a step, do the work, pass context to the next agent. SQLite tracks state. Cron keeps it moving.\n\n### Minimal by design\n\nYAML + SQLite + cron. That's it. No Redis, no Kafka, no container orchestrator. Antfarm is a TypeScript CLI with zero external dependencies. It runs wherever OpenClaw runs.\n\n### Built on the Ralph loop\n\n\u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fsnarktank\u002Fralph\u002Fmain\u002Fralph.webp\" alt=\"Ralph\" width=\"100\">\n\nEach agent runs in a fresh session with clean context. Memory persists through git history and progress files — the same autonomous loop pattern from [Ralph](https:\u002F\u002Fgithub.com\u002Fsnarktank\u002Fralph), scaled to multi-agent workflows.\n\n---\n\n## Quick Example\n\n```bash\n$ antfarm workflow install feature-dev\n✓ Installed workflow: feature-dev\n\n$ antfarm workflow run feature-dev \"Add user authentication with OAuth\"\nRun: a1fdf573\nWorkflow: feature-dev\nStatus: running\n\n$ antfarm workflow status \"OAuth\"\nRun: a1fdf573\nWorkflow: feature-dev\nSteps:\n  [done   ] plan (planner)\n  [done   ] setup (setup)\n  [running] implement (developer)  Stories: 3\u002F7 done\n  [pending] verify (verifier)\n  [pending] test (tester)\n  [pending] pr (developer)\n  [pending] review (reviewer)\n```\n\n---\n\n## Build Your Own\n\nThe bundled workflows are starting points. Define your own agents, steps, retry logic, and verification gates in plain YAML and Markdown. If you can write a prompt, you can build a workflow.\n\n```yaml\nid: my-workflow\nname: My Custom Workflow\nagents:\n  - id: researcher\n    name: Researcher\n    workspace:\n      files:\n        AGENTS.md: agents\u002Fresearcher\u002FAGENTS.md\n\nsteps:\n  - id: research\n    agent: researcher\n    input: |\n      Research {{task}} and report findings.\n      Reply with STATUS: done and FINDINGS: ...\n    expects: \"STATUS: done\"\n```\n\nFull guide: [docs\u002Fcreating-workflows.md](docs\u002Fcreating-workflows.md)\n\n---\n\n## Security\n\nYou're installing agent teams that run code on your machine. We take that seriously.\n\n- **Curated repo only** — Antfarm only installs workflows from the official [snarktank\u002Fantfarm](https:\u002F\u002Fgithub.com\u002Fsnarktank\u002Fantfarm) repository. No arbitrary remote sources.\n- **Reviewed for prompt injection** — Every workflow is reviewed for prompt injection attacks and malicious agent files before merging.\n- **Community contributions welcome** — Want to add a workflow? Submit a PR. All submissions go through careful security review before they ship.\n- **Transparent by default** — Every workflow is plain YAML and Markdown. You can read exactly what each agent will do before you install it.\n\n---\n\n## Dashboard\n\nMonitor runs, track step progress, and view agent output in real time.\n\n![Antfarm dashboard](https:\u002F\u002Fraw.githubusercontent.com\u002Fsnarktank\u002Fantfarm\u002Fmain\u002Fassets\u002Fdashboard-screenshot.png)\n\n![Antfarm dashboard detail](https:\u002F\u002Fraw.githubusercontent.com\u002Fsnarktank\u002Fantfarm\u002Fmain\u002Fassets\u002Fdashboard-detail-screenshot.png)\n\n```bash\nantfarm dashboard              # Start on port 3333\nantfarm dashboard stop         # Stop\nantfarm dashboard status       # Check status\n```\n\n---\n\n## Commands\n\n### Lifecycle\n\n| Command | Description |\n|---------|-------------|\n| `antfarm install` | Install all bundled workflows |\n| `antfarm uninstall [--force]` | Full teardown (agents, crons, DB) |\n\n### Workflows\n\n| Command | Description |\n|---------|-------------|\n| `antfarm workflow run \u003Cid> \u003Ctask>` | Start a run |\n| `antfarm workflow status \u003Cquery>` | Check run status |\n| `antfarm workflow runs` | List all runs |\n| `antfarm workflow resume \u003Crun-id>` | Resume a failed run |\n| `antfarm workflow list` | List available workflows |\n| `antfarm workflow install \u003Cid>` | Install a single workflow |\n| `antfarm workflow uninstall \u003Cid>` | Remove a single workflow |\n\n### Management\n\n| Command | Description |\n|---------|-------------|\n| `antfarm dashboard` | Start the web dashboard |\n| `antfarm logs [\u003Clines>]` | View recent log entries |\n\n---\n\n## Requirements\n\n- Node.js >= 22\n- [OpenClaw](https:\u002F\u002Fgithub.com\u002Fopenclaw\u002Fopenclaw) **v2026.2.9+** running on the host\n  - Antfarm uses cron jobs for workflow orchestration. Older OpenClaw versions may not expose the cron tool via `\u002Ftools\u002Finvoke`. Antfarm will automatically fall back to the `openclaw` CLI, but keeping OpenClaw up to date is recommended: `npm update -g openclaw`\n- `gh` CLI for PR creation steps\n\n---\n\n## License\n\n[MIT](LICENSE)\n\n---\n\n\u003Cp align=\"center\">Part of the \u003Ca href=\"https:\u002F\u002Fdocs.openclaw.ai\">OpenClaw\u003C\u002Fa> ecosystem · Built by \u003Ca href=\"https:\u002F\u002Fryancarson.com\">Ryan Carson\u003C\u002Fa>\u003C\u002Fp>\n","Antfarm 是一个基于 OpenClaw 的工具，允许用户通过一条命令构建由多个 AI 代理组成的团队。其核心功能包括提供一系列预定义的工作流，如功能开发、安全审计和错误修复等，每个工作流都由专门的代理（如规划者、开发者、验证者等）协同完成任务。Antfarm 采用 TypeScript 开发，具有确定性工作流程、代理间相互验证以及自动重试机制等特点，确保了过程的可靠性和一致性。适用于需要快速搭建自动化软件开发或维护流程而无需额外基础设施支持的场景。",2,"2026-06-11 03:51:06","high_star"]