{
  "markdown": "# AgiFlow AI Plugin\n\nOfficial AgiFlow plugin for AI clients. Drive AgiFlow project management, including planning,\ngrooming, execution, and review, directly from your AI tool.\n\nWorks with **ChatGPT**, **Claude Code**, **Codex**, **Cursor**, **Antigravity**, and **Gemini CLI**.\n\n## Installation\n\nThis repo is a self-contained, multi-client plugin bundle. Until it is published to each client's\nmarketplace, load it as a local plugin directory.\n\n### Claude Code\n\n```bash\ngit clone <your-remote>/agiflow-ai-plugin\nclaude --plugin-dir ./agiflow-ai-plugin\n```\n\nThe bundled `.mcp.json` wires the AgiFlow MCP server automatically. Use `/mcp` inside Claude Code to\ncheck the connection.\n\n### Antigravity (Google)\n\nPlace the plugin folder in one of Antigravity's plugin locations, then restart:\n\n```bash\n# Workspace-level (this project only)\nmkdir -p .agents/plugins && cp -R /path/to/agiflow-ai-plugin .agents/plugins/\n\n# Global (all workspaces)\nmkdir -p ~/.gemini/config/plugins && cp -R /path/to/agiflow-ai-plugin ~/.gemini/config/plugins/\n```\n\nAntigravity reads the root `plugin.json` marker, the `skills/`, and `mcp_config.json` automatically.\n\n### Cursor\n\nAdd manually in **Cursor Settings → MCP / Plugins**, pointing at this folder. Cursor's stable surface\nis MCP config. The bundled `.mcp.json` provides it.\n\n### Codex\n\nAdd the AgiFlow plugin marketplace, then install the plugin from that marketplace:\n\n```bash\ncodex plugin marketplace add AgiFlow/ai-plugin\ncodex plugin add agiflow-ai-plugin@agiflow\n```\n\nFor local development, point Codex at this checkout as a marketplace root:\n\n```bash\ncodex plugin marketplace add ./agiflow-ai-plugin\ncodex plugin add agiflow-ai-plugin@agiflow\n```\n\n### Gemini CLI\n\n```bash\ngemini extensions install <your-remote>/agiflow-ai-plugin\n```\n\nThe bundled `gemini-extension.json` connects the AgiFlow MCP server via `mcp-remote`.\n\n### ChatGPT\n\nOpenAI Platform plugins are submitted from the production AgiFlow MCP server. This public repository\nprovides the reusable Agent Skills used by that plugin, but does not contain Platform dashboard IDs,\nsubmission evidence, reviewer credentials, or other private operational data.\n\nThe ChatGPT-safe workflow set is:\n\n- `getting-started`\n- `project-plan`\n- `refine-task`\n- `backlog-grooming`\n- `daily-standup`\n- `triage`\n\nThe coding-agent workflows `orchestrate`, `run-task`, `run-work`, and `review-work` remain available\nto coding clients but are not part of the ChatGPT submission bundle.\n\n## How to develop\n\n```bash\ngit clone <your-remote>/agiflow-ai-plugin\nclaude --plugin-dir ./agiflow-ai-plugin\n```\n\n- Add new workflow instructions under `skills/<name>/SKILL.md`.\n- Keep shared guidance in `references/` (e.g. `references/agiflow-agents.md`).\n- See `references/plugin-types.md` for per-client manifest notes.\n\n## Features\n\nThis plugin connects to the AgiFlow MCP server (`https://agiflow.io/api/v1/mcp`) and exposes AgiFlow\ntools across these categories:\n\n- **Projects**: create, inspect, and update projects and their statuses\n- **Tasks**: create, list, get, update, reorder, and batch-create tasks\n- **Work units**: group tasks into deliverable features or epics and track progress\n- **Workflows**: acquire and release locks and coordinate multi-agent runs\n- **Members**: list and assign agent members to work\n- **Comments**: document decisions and progress on tasks\n- **Vault**: read and set scoped configuration entries\n\n### Bundled skills\n\nThe plugin ships 10 workflow skills that mirror AgiFlow's scrum pipeline. Your AI client loads them on\ndemand when your request matches their description. You generally do not invoke them by name:\n\n| Skill | Phase | Use it to |\n| --- | --- | --- |\n| `getting-started` | orient | get coached on where to start and which workflow fits |\n| `project-plan` | Planning | break requirements into vertical-slice tasks (Planning status) |\n| `refine-task` | Planning | turn a vague task into an autonomous-ready spec |\n| `backlog-grooming` | Planning → Todo | verify, prioritize, and promote tasks into work units |\n| `run-work` | Todo → Done | execute a whole work unit end-to-end in one session |\n| `run-task` | Todo → Done | execute a single task through to Review |\n| `review-work` | Review | verify acceptance criteria and file follow-ups |\n| `triage` | diagnose | classify project issues by severity and recommend actions |\n| `daily-standup` | report | a read-only pulse of done / in-progress / blocked / next |\n| `orchestrate` | dispatch | route the highest-priority ready work to agents |\n\nShared guidelines (status model, transitions, tags, work-unit sizing) live in\n[`references/agiflow-agents.md`](references/agiflow-agents.md).\n\n## Example usage\n\n```\n> Plan a feature: add per-user notification preferences\n> Groom the backlog and promote the ready tasks to Todo\n\n> Run task DXX-2\n> Execute the checkout work unit end-to-end\n\n> Review the auth work unit against its acceptance criteria\n> Give me a daily standup for this project\n\n> Why is this project stuck?\n> What should an agent pick up next?\n```\n\n## Self-hosted\n\nFor a self-hosted AgiFlow instance, point the MCP wiring at your endpoint via the\n`AGIFLOW_AI_PLUGIN_MCP_URL` environment variable (consumed by `gemini-extension.json`):\n\n```bash\nexport AGIFLOW_AI_PLUGIN_MCP_URL=\"https://mcp.your-agiflow-instance.com/api/v1/mcp\"\n```\n\nFor other clients, edit the server URL in `.mcp.json`, `mcp.json`, and `mcp_config.json`.\n\n## Documentation\n\n- AgiFlow: https://agiflow.io\n- Plugin client compatibility: [`references/plugin-types.md`](references/plugin-types.md)\n\n## License\n\nMIT\n",
  "bytes": 5558,
  "sha": "2cc2cb387387af3e4d9581759302ec5a2fef9ce8910daac910ac8e43e1c0f609",
  "repo_slug": "agiflow/ai-plugin",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_agiflow_ai_plugin_52d1e08a/readme"
}