{"slug":"agent-zero","name":"Agent Zero","vendor":"Agent Zero, s.r.o.","tagline":"Open-source agent framework that gives the model a real Linux desktop, browser and terminal inside Docker","url":"https://agentboards.org/agent/agent-zero","website":"https://agent-zero.ai","docs":null,"repo":"https://github.com/agent0ai/agent-zero","category":"framework","execution":["local","sandbox"],"license":"MIT","open_source":true,"pricing":{"model":"byok","summary":"Free and open source; you bring your own provider keys or run local models with Ollama","from_usd":null,"free_tier":true,"byok":true},"models":{"backbone":["any"],"bring_your_own_model":true,"local_models":true},"protocols":{"mcp_client":true,"mcp_server":true,"openapi":false},"capabilities":{"terminal_exec":true,"browser":true,"multi_file_edit":true,"git_ops":true,"docker_sandbox":true,"multi_agent":true,"headless_ci":false},"context_window":null,"platforms":["macos","linux","windows"],"install":[{"label":"Docker","command":"docker run -p 80:80 -v a0_usr:/a0/usr agent0ai/agent-zero"}],"benchmarks":[],"tags":["framework","open-source","docker","computer-use","mcp","multi-agent"],"github_stars":19357,"metrics":{"stars":19357,"pushed_at":"2026-09-30T18:48:34Z","archived":false,"open_issues":134,"latest_version":"v2.13","latest_version_at":"2026-09-23T18:47:44Z","latest_version_url":"https://github.com/agent0ai/agent-zero/releases/tag/v2.13","version_source":"github","checked_at":"2026-10-02T05:04:05.335Z"},"sources":{"license":"https://github.com/agent0ai/agent-zero","capabilities":"https://github.com/agent0ai/agent-zero","models":"https://github.com/agent0ai/agent-zero","protocols":"https://github.com/agent0ai/agent-zero","install":"https://github.com/agent0ai/agent-zero"},"verified_at":"2026-09-04","adoption":{"score":6,"signals":{"github_stars":19290,"hn_mentions_1y":6},"measured_at":"2026-09-28"},"scores":{"panel":6.3,"adoption":6,"spread":3,"facts":8.3,"board":6.6,"dimensions":{"reliability":5.5,"usefulness":6.7,"cost":7,"longevity":6},"per_persona":{"amigo":6.8,"critico":6.3,"profesor":6.5,"inversora":5.3,"jefa":5,"hacker":8},"community":null},"rank":null,"category_rank":22,"panel_reviews":[{"persona":"juez","persona_name":"El Juez","ai_generated":true,"verdict":"El Hacker scores it eight for MIT, LiteLLM and MCP at both ends; El Crítico points at the same editability and calls it a blast radius that includes the rules.","scores":null,"evidence":[]},{"persona":"amigo","persona_name":"El Amigo","ai_generated":true,"verdict":"Pick Agent Zero when the job is driving real desktop applications; pick OpenHands when the job is a git repository and a test suite.","scores":{"reliability":6,"usefulness":7,"cost":8,"longevity":6},"evidence":["https://agent-zero.ai","https://github.com/agent0ai/agent-zero"]},{"persona":"critico","persona_name":"El Crítico","ai_generated":true,"verdict":"Every prompt, tool and config file is editable by design, and the agent has read and write access to the same files, so the rules it follows are inside its own blast radius.","scores":{"reliability":5,"usefulness":6,"cost":8,"longevity":6},"evidence":["https://github.com/agent0ai/agent-zero","https://agent-zero.ai"]},{"persona":"profesor","persona_name":"El Profesor","ai_generated":true,"verdict":"Context arrives through a DOM-annotated browser rather than raw HTML, and work is delegated to subordinate agents that each keep an isolated context.","scores":{"reliability":6,"usefulness":7,"cost":6,"longevity":7},"evidence":["https://github.com/agent0ai/agent-zero"]},{"persona":"inversora","persona_name":"La Inversora","ai_generated":true,"verdict":"Agent Zero, s.r.o. is a small European entity holding a 19,000-star project with no paid tier, sitting in the one category the model labs are entering themselves.","scores":{"reliability":5,"usefulness":6,"cost":5,"longevity":5},"evidence":["https://agent-zero.ai","https://github.com/agent0ai/agent-zero"]},{"persona":"jefa","persona_name":"La Jefa","ai_generated":true,"verdict":"No seat cost, but a container per engineer with a full desktop inside it is a compute line, and there is no headless mode, so nothing runs in our pipelines.","scores":{"reliability":4,"usefulness":5,"cost":6,"longevity":5},"evidence":["https://github.com/agent0ai/agent-zero","https://agent-zero.ai"]},{"persona":"hacker","persona_name":"El Hacker","ai_generated":true,"verdict":"MIT, one docker run with an a0_usr volume, LiteLLM in front of every provider, Ollama for local weights, and it works as an MCP client and an MCP server.","scores":{"reliability":7,"usefulness":9,"cost":9,"longevity":7},"evidence":["https://github.com/agent0ai/agent-zero","https://agent-zero.ai"]},{"persona":"comediante","persona_name":"El Comediante","ai_generated":true,"verdict":"It can cowork with you on LibreOffice documents, which is the most 2003 sentence ever written about an autonomous agent.","scores":null,"evidence":[]}]}