{"user":"vxcontrol","name":"pentagi","namespace":"vxcontrol","repository_type":"image","status":1,"status_description":"active","description":"✨ A fully autonomous AI Agent capable of performing complex penetration testing tasks","is_private":false,"is_automated":false,"star_count":10,"pull_count":350989,"last_updated":"2026-10-05T07:43:25.815075Z","last_modified":"2025-06-25T19:24:19.95115Z","date_registered":"2025-01-04T09:28:46.980035Z","collaborator_count":0,"affiliation":null,"hub_user":"vxcontrol","has_starred":false,"full_description":"# PentAGI - AI-Powered Penetration Testing\n\n\u003e [![Discord](https://img.shields.io/badge/Discord-7289DA?logo=discord\u0026logoColor=white)](https://discord.gg/2xrMh7qX6m)⠀[![Telegram](https://img.shields.io/badge/Telegram-2CA5E0?logo=telegram\u0026logoColor=white)](https://t.me/+Ka9i6CNwe71hMWQy)\u003c/br\u003e\n\u003e **Join the Community!** Connect with security researchers, AI enthusiasts, and fellow ethical hackers. Get support, share insights, and stay updated with the latest PentAGI developments.\n\n**P**enetration testing **A**rtificial **G**eneral **I**ntelligence - A fully autonomous AI Agent capable of performing complex penetration testing tasks using terminal, browser, editor, and external search systems.\n\n## Key Features\n\n- 🛡️ **Secure \u0026 Isolated**: All operations are performed in a sandboxed Docker environment\n- 🤖 **Fully Autonomous**: AI-powered agent that automatically determines and executes testing steps\n- 🔬 **Professional Tools**: Built-in suite of 20+ security tools (nmap, metasploit, sqlmap, etc.)\n- 🧠 **Smart Memory**: Long-term storage of research results and successful approaches\n- 🔍 **Web Intelligence**: Built-in browser for gathering latest information\n- 🔎 **External Search**: Integration with Tavily, Traversaal, and Google Custom Search\n- 👥 **Team of Specialists**: Specialized AI agents for research, development, and infrastructure\n- 📊 **Monitoring**: Integration with Grafana/Prometheus for real-time observation\n- 📝 **Reporting**: Detailed vulnerability reports with exploitation guides\n\n## Quick Start\n\n`1.` Create a working directory:\n```bash\nmkdir pentagi \u0026\u0026 cd pentagi\n```\n\n`2.` Download configuration files:\n```bash\ncurl -O https://raw.githubusercontent.com/vxcontrol/pentagi/master/.env.example\ncurl -O https://raw.githubusercontent.com/vxcontrol/pentagi/master/docker-compose.yml\n```\n`3.` Configure environment:\n```bash\ncp .env.example .env\n# Edit .env file with your API keys and settings\n```\n\n`4.` Run PentAGI:\n```bash\ndocker compose up -d\n```\n\n`5.` Access the UI at [https://localhost:8443](https://localhost:8443) (default is `admin@pentagi.com` / `admin`)\n\n## Required Environment Variables\n\n- `OPEN_AI_KEY` or `ANTHROPIC_API_KEY`: At least one LLM provider key\n\n## Optional Environment Variables\n\n- `GOOGLE_API_KEY` \u0026 `GOOGLE_CX_KEY`: For Google search integration\n- `TAVILY_API_KEY`: For Tavily search integration\n- `TRAVERSAAL_API_KEY`: For Traversaal search integration\n- `PROXY_URL`: Global proxy for external services\n- `PUBLIC_URL`: Public URL of your server\n- `PENTAGI_POSTGRES_USER` \u0026 `PENTAGI_POSTGRES_PASSWORD`: PostgreSQL credentials\n- `SERVER_SSL_CRT` \u0026 `SERVER_SSL_KEY`: Custom SSL certificate paths\n\n## System Requirements\n\n- Docker Engine 24.0+\n- Docker Compose v2.0+\n- 4GB RAM minimum\n- 20GB disk space\n- Internet access\n\n## Additional Features\n\n### Monitoring Stack\n- OpenTelemetry for data collection\n- Grafana for visualization (port 3000)\n- VictoriaMetrics for metrics\n- Jaeger for tracing\n- Loki for logs\n\nQuick setup:\n```bash\n# 1. Enable OpenTelemetry integration\necho \"OTEL_HOST=otelcol:8148\" \u003e\u003e .env\n\n# 2. Download and start the monitoring stack\ncurl -O https://raw.githubusercontent.com/vxcontrol/pentagi/master/docker-compose-observability.yml\ndocker compose -f docker-compose.yml -f docker-compose-observability.yml up -d\n```\n\nAccess Grafana at [http://localhost:3000](http://localhost:3000)\n\n[Learn more about monitoring setup →](https://github.com/vxcontrol/pentagi#monitoring-and-observability)\n\n### Analytics Platform\n- Langfuse for LLM observability (port 4000)\n- ClickHouse for analytics\n- Redis for caching\n- MinIO for storage\n\nQuick setup:\n```bash\n# 1. Enable Langfuse integration\necho \"LANGFUSE_BASE_URL=http://langfuse-web:3000\" \u003e\u003e .env\n\n# 2. Download and start the analytics stack\ncurl -O https://raw.githubusercontent.com/vxcontrol/pentagi/master/docker-compose-langfuse.yml\ndocker compose -f docker-compose.yml -f docker-compose-langfuse.yml up -d\n```\n\nAccess Langfuse at [http://localhost:4000](http://localhost:4000)\n\n[Learn more about analytics setup →](https://github.com/vxcontrol/pentagi#langfuse-integration)\n\n\u003e **Note**: For production use, you should properly configure security settings and credentials in the `.env` file. See the full documentation for details.\n\n\u003e **Tip**: To run both monitoring and analytics stacks together:\n\u003e ```bash\n\u003e docker compose -f docker-compose.yml -f docker-compose-langfuse.yml -f docker-compose-observability.yml up -d\n\u003e ```\n\n## Documentation\n\nFull documentation available at:\n- GitHub Repository: [vxcontrol/pentagi](https://github.com/vxcontrol/pentagi)\n- Documentation: [docs.pentagi.com](https://docs.pentagi.com)\n\n## Support\n\n- Report issues on [GitHub](https://github.com/vxcontrol/pentagi/issues)\n- Join discussions on [GitHub Discussions](https://github.com/vxcontrol/pentagi/discussions)\n- Mail to [info@pentagi.com](mailto:info@pentagi.com)\n\n## License\n\nCopyright (c) PentAGI Development Team. [MIT License](https://github.com/vxcontrol/pentagi/blob/master/LICENSE)","permissions":{"read":true,"write":false,"admin":false},"media_types":["application/vnd.oci.image.index.v1+json"],"content_types":["image"],"categories":[{"name":"Security","slug":"security"},{"name":"Machine learning \u0026 AI","slug":"machine-learning-and-ai"},{"name":"Developer tools","slug":"developer-tools"}],"immutable_tags_settings":{"enabled":false,"rules":[".*"]},"storage_size":47606474409,"source":null}