{"data":{"slug":"cordum-io-cordum","name":"cordum","tagline":"The action firewall for AI agents, enforcing policy and human approval before risky operations.","github_url":"https://github.com/cordum-io/cordum","owner":"cordum-io","repo":"cordum","owner_avatar_url":"https://avatars.githubusercontent.com/u/248097825?v=4","primary_language":"Go","stars":507,"forks":33,"topics":["agent-framework","agentic-ai","ai-agent","ai-governance","ai-orchestration","ai-safety","audit-trail","autonomous-agents","control-plane","devops","governance","human-in-the-loop","llm","llm-agents","mcp","model-context-protocol","nats","policy-engine","safety-kernel","workflow-engine"],"archived":false,"github_pushed_at":"2026-09-17T22:49:32+00:00","maintenance_label":"Very active","stars_delta_30d":13,"url":"https://www.graphcanon.com/tools/cordum-io-cordum","markdown_url":"https://www.graphcanon.com/tools/cordum-io-cordum.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/cordum-io-cordum","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=cordum-io-cordum","description":"The action firewall for AI agents. Enforce policy and human approval before risky tool calls, shell commands, workflows, and production changes, with auditable evidence.","homepage_url":null,"license":"Other","open_issues":30,"watchers":131,"ai_summary":"Cordum is a framework for governing AI agents, ensuring that actions are approved by humans and audited for compliance. It includes a dashboard, workflow engine, and safety kernel.","readme_excerpt":"### Option A: Install from source\n\nOne command stands up the full stack — API gateway, scheduler, safety\nkernel, workflow engine, context engine, dashboard, NATS, and\nTLS-secured Redis — with auto-generated secrets, auto-provisioned\ncertificates, and a post-deploy smoke test that exercises a real\napproval workflow:\n\n```bash\ngit clone https://github.com/cordum-io/cordum.git\ncd cordum\n./tools/scripts/quickstart.sh\n```\n\n**Prerequisites:** Docker Desktop v4+ (or Engine v20.10+ with Compose v2,\n≥ 4 GB RAM allocated), Go 1.26.3+ (for first-run cert generation), and\n`curl`. On Windows use MSYS2 / Git Bash / WSL.\n\n**What you get at the end:**\n- Dashboard at http://localhost:8082 — log in as `admin` / `ChangeMe123!`\n  (the default dev password, also saved to `.env` as `CORDUM_ADMIN_PASSWORD`;\n  change it before exposing the stack).\n- Gateway at http://localhost:8081 with a generated `CORDUM_API_KEY` in\n  `.env`.\n- TLS CA, server, and client keypairs under `./certs/`.\n- A working approval-gate workflow proven by the built-in\n  `platform_smoke.sh` run.\n\nFull walkthrough, platform notes, and troubleshooting:\n[docs/quickstart.md](docs/quickstart.md).\n\n---\n\n## License\n\nLicensed under [Business Source License 1.1 (BUSL-1.1)](LICENSE).\n\n- **Self-host and use internally**: Permitted\n- **Modify and contribute back**: Permitted\n- **Offer as a competing hosted service**: Not permitted\n- **Change Date**: January 1, 2029 — automatically converts to Apache License 2.0\n\nSee [LICENSE](LICENSE) for full terms.\n\n---","github_created_at":"2026-01-11T16:56:02+00:00","created_at":"2026-07-11T23:13:41.58787+00:00","updated_at":"2026-09-20T04:30:09.656217+00:00","categories":[{"slug":"ai-agents","name":"AI Agents","url":"https://www.graphcanon.com/categories/ai-agents","markdown_url":"https://www.graphcanon.com/categories/ai-agents.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/ai-agents"},{"slug":"evaluation-observability","name":"Evaluation & Observability","url":"https://www.graphcanon.com/categories/evaluation-observability","markdown_url":"https://www.graphcanon.com/categories/evaluation-observability.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/evaluation-observability"}],"tags":[{"slug":"agent-framework","name":"agent-framework"},{"slug":"ai-governance","name":"ai-governance"},{"slug":"audit-trail","name":"audit-trail"},{"slug":"human-in-the-loop","name":"human-in-the-loop"},{"slug":"policy-engine","name":"policy-engine"},{"slug":"safety-kernel","name":"safety-kernel"},{"slug":"workflow-engine","name":"workflow-engine"}],"trust":{"provenance":{"is_fork":false,"github_id":1132275655,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-18T08:21:26.675Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":0,"days_since_push":0,"last_release_at":"2026-06-03T14:11:53Z","stars_delta_30d":13,"open_issues_delta_30d":14},"security_summary":{"status":"findings","scanner":"osv@v1","low_count":13,"high_count":0,"last_scan_at":"2026-09-18T08:21:28.328Z","medium_count":0,"scan_profile":"deps","critical_count":0}},"capability_facts":{"mcp":{"source":"repo_scan","observed_at":"2026-09-18T08:21:25.552Z","server_manifest":false},"scan":{"source":"repo_scan","observed_at":"2026-09-18T08:21:25.552Z"},"deploy":{"source":"dockerfile:Dockerfile","self_host":true,"observed_at":"2026-09-18T08:21:25.552Z","managed_saas":false},"languages":{"value":["go","javascript"],"source":"github.language+package.json","observed_at":"2026-09-18T08:21:25.552Z"},"has_docker":{"value":true,"source":"dockerfile:Dockerfile","observed_at":"2026-09-18T08:21:25.552Z"},"license_spdx":{"value":"Other","source":"github.license","observed_at":"2026-09-18T08:21:25.552Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["When you need a framework that ensures human approval before AI agents execute risky operations, such as shell commands or production changes.","If your organization requires a comprehensive audit trail for AI actions, including a dashboard and workflow engine for tracking and compliance.","When deploying AI agents in environments where governance and safety are critical, and you need a control plane that integrates with a NATS messaging system and TLS-secured Redis.","For scenarios where you can self-host the solution and do not plan to offer it as a competing hosted service before January 1, 2029."],"when_not_to_use":["If you are looking for a solution that can be offered as a competing hosted service before January 1, 2029, as the current license restricts this use case.","When your deployment environment does not meet the prerequisites for running Cordum, such as Docker Desktop v4+, Go 1.26.3+, and at least 4 GB of RAM.","If your organization does not require a full stack setup including an API gateway, scheduler, safety kernel, workflow engine, context engine, dashboard, NATS, and TLS-secured Redis.","When you prefer a tool that does not enforce a human-in-the-loop approval process for AI actions, and you are comfortable with less stringent governance measures."],"source":"enrich:decision_facts","observed_at":"2026-09-18T11:00:30.148Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"Cordum is an action firewall for AI agents, designed to enforce policy and human approval before risky operations, with a focus on auditability and compliance."}]}}