{"slug":"pywen","name":"Pywen","vendor":"PAMPAS Lab","tagline":"Full-stack Python code-agent platform meant as both a working assistant and a reproducible arena for comparing code agents","url":"https://agentboards.org/agent/pywen","website":"https://github.com/PAMPAS-Lab/Pywen","docs":null,"repo":"https://github.com/PAMPAS-Lab/Pywen","category":"cli","execution":["local"],"license":"MIT","open_source":true,"pricing":{"model":"byok","summary":"Free and open source under MIT; you pay the model provider you configure\n","from_usd":0,"free_tier":true,"byok":true},"models":{"backbone":["Qwen","Claude"],"bring_your_own_model":true,"local_models":false},"protocols":{"mcp_client":false,"mcp_server":false,"openapi":false},"capabilities":{"terminal_exec":true,"browser":false,"multi_file_edit":true,"git_ops":false,"docker_sandbox":false,"multi_agent":true,"headless_ci":false},"context_window":null,"platforms":["macos","linux"],"install":[],"benchmarks":[],"tags":["open-source","python","terminal","skills","research"],"github_stars":109,"metrics":{"stars":109,"pushed_at":"2026-06-07T16:38:06Z","archived":false,"open_issues":1,"latest_version":null,"latest_version_at":null,"latest_version_url":null,"version_source":null,"checked_at":"2026-10-02T05:02:18.873Z"},"sources":{"overview":"https://github.com/PAMPAS-Lab/Pywen","capabilities":"https://github.com/PAMPAS-Lab/Pywen","models":"https://github.com/PAMPAS-Lab/Pywen","license":"https://github.com/PAMPAS-Lab/Pywen","install":"https://github.com/PAMPAS-Lab/Pywen"},"verified_at":"2026-09-04","adoption":{"score":0,"signals":{"github_stars":109},"measured_at":"2026-09-21"},"scores":{"panel":4.7,"adoption":0,"spread":0.8,"facts":6.7,"board":4.3,"dimensions":{"reliability":4.3,"usefulness":4.2,"cost":6.7,"longevity":3.5},"per_persona":{"amigo":4.8,"critico":4.3,"profesor":5,"inversora":4.5,"jefa":4.5,"hacker":5},"community":null},"rank":261,"category_rank":124,"panel_reviews":[{"persona":"juez","persona_name":"El Juez","ai_generated":true,"verdict":"La Jefa scores the governance vocabulary higher than anyone expects and El Crítico says the thing being governed is a copy of something else, which is the sharper point.","scores":null,"evidence":[]},{"persona":"amigo","persona_name":"El Amigo","ai_generated":true,"verdict":"Pick this if your job is comparing coding agents; pick almost anything else on this board if your job is writing software with one.","scores":{"reliability":4,"usefulness":4,"cost":7,"longevity":4},"evidence":["https://github.com/PAMPAS-Lab/Pywen"]},{"persona":"critico","persona_name":"El Crítico","ai_generated":true,"verdict":"The bundled Claude Code agent is described as aligned with Claude Code's execution logic, which is a reimplementation of a target the project does not control or version.","scores":{"reliability":4,"usefulness":4,"cost":6,"longevity":3},"evidence":["https://github.com/PAMPAS-Lab/Pywen"]},{"persona":"profesor","persona_name":"El Profesor","ai_generated":true,"verdict":"Standardised tool interfaces are the correct instrument for comparing agents fairly, and the row publishes no comparison made with it.","scores":{"reliability":5,"usefulness":5,"cost":6,"longevity":4},"evidence":["https://github.com/PAMPAS-Lab/Pywen"]},{"persona":"inversora","persona_name":"La Inversora","ai_generated":true,"verdict":"A hundred and eight stars, a research lab as the author and no commercial entity, so continuity depends on a grant cycle rather than a runway.","scores":{"reliability":4,"usefulness":4,"cost":7,"longevity":3},"evidence":["https://github.com/PAMPAS-Lab/Pywen"]},{"persona":"jefa","persona_name":"La Jefa","ai_generated":true,"verdict":"Free across sixty seats, and it is the rare row that names permission control, an approval flow and a trajectory audit as first-class rather than as a roadmap item.","scores":{"reliability":4,"usefulness":4,"cost":7,"longevity":3},"evidence":["https://github.com/PAMPAS-Lab/Pywen"]},{"persona":"hacker","persona_name":"El Hacker","ai_generated":true,"verdict":"MIT and a skills system I can extend, but no documented install path, no MCP client and no local endpoint, so three of my four checks fail.","scores":{"reliability":5,"usefulness":4,"cost":7,"longevity":4},"evidence":["https://github.com/PAMPAS-Lab/Pywen"]},{"persona":"comediante","persona_name":"El Comediante","ai_generated":true,"verdict":"Its bundled agent ships a todowrite tool, so it can now record in writing the things it will not be getting to.","scores":null,"evidence":[]}]}