{"slug":"agently","name":"Agently","vendor":"AgentEra","tagline":"Python AI application runtime with structured outputs, observable actions, runtime skills, MCP tools and restart-safe workflows","url":"https://agentboards.org/agent/agently","website":"https://agently.tech","docs":"https://agently.tech/docs","repo":"https://github.com/AgentEra/Agently","category":"framework","execution":["local","sandbox"],"license":"Apache-2.0","open_source":true,"pricing":{"model":"byok","summary":"Free and open source under Apache-2.0 on PyPI; you configure your own provider endpoints and keys","from_usd":0,"free_tier":true,"byok":true},"models":{"backbone":["DeepSeek","OpenAI","Ollama","any OpenAI-compatible endpoint"],"bring_your_own_model":true,"local_models":true},"protocols":{"mcp_client":true,"mcp_server":false,"openapi":false},"capabilities":{"terminal_exec":true,"browser":true,"multi_file_edit":false,"git_ops":false,"docker_sandbox":false,"multi_agent":false,"headless_ci":false},"context_window":null,"platforms":["macos","linux","windows"],"install":[{"label":"pip","command":"pip install -U agently"}],"benchmarks":[],"tags":["open-source","python","structured-output","mcp","workflows","sandbox","observability"],"github_stars":1655,"metrics":{"stars":1655,"pushed_at":"2026-10-02T00:40:59Z","archived":false,"open_issues":16,"latest_version":"4.1.4.9","latest_version_at":"2026-10-01T15:36:38.583263Z","latest_version_url":"https://pypi.org/project/agently/","version_source":"pypi","checked_at":"2026-10-02T05:04:09.652Z"},"sources":{"readme":"https://github.com/AgentEra/Agently","docs":"https://agently.tech/docs","install":"https://github.com/AgentEra/Agently","models":"https://github.com/AgentEra/Agently","protocols":"https://github.com/AgentEra/Agently/blob/main/docs/en/actions/mcp.md","license":"https://github.com/AgentEra/Agently/blob/main/LICENSE","capabilities":"https://github.com/AgentEra/Agently/blob/main/docs/en/actions/execution-environment.md"},"verified_at":null,"adoption":{"score":2.5,"signals":{"pypi_weekly":850,"github_stars":1652,"hn_mentions_1y":1},"measured_at":"2026-09-21"},"scores":{"panel":6.3,"adoption":2.5,"spread":2,"facts":5.8,"board":5.7,"dimensions":{"reliability":5.7,"usefulness":5.8,"cost":8.2,"longevity":5.3},"per_persona":{"amigo":6.3,"critico":5.8,"profesor":6.8,"inversora":5.8,"jefa":5.5,"hacker":7.5},"community":null},"rank":null,"category_rank":62,"panel_reviews":[{"persona":"juez","persona_name":"El Juez","ai_generated":true,"verdict":"El Hacker's 8 and El Crítico's 4 for reliability sit on the same page of documentation: the isolation mechanisms ship inactive, and inactive is the default.","scores":null,"evidence":[]},{"persona":"amigo","persona_name":"El Amigo","ai_generated":true,"verdict":"Pick Agently if you want structured records out of every run rather than prose; pick PydanticAI when the typed output is the whole reason you are here.","scores":{"reliability":6,"usefulness":6,"cost":8,"longevity":5},"evidence":["https://agently.tech/docs","https://github.com/AgentEra/Agently"]},{"persona":"critico","persona_name":"El Crítico","ai_generated":true,"verdict":"The gVisor, Seatbelt and Landlock code-execution candidates ship inactive and only probe their mechanism when selected, so nothing confines the default path.","scores":{"reliability":4,"usefulness":6,"cost":8,"longevity":5},"evidence":["https://github.com/AgentEra/Agently/blob/main/docs/en/actions/execution-environment.md","https://github.com/AgentEra/Agently"]},{"persona":"profesor","persona_name":"El Profesor","ai_generated":true,"verdict":"An AgentExecution owns exactly one run's prompt, strategy, actions and skills, which gives the unit of work a boundary most frameworks leave implicit.","scores":{"reliability":7,"usefulness":6,"cost":8,"longevity":6},"evidence":["https://agently.tech/docs","https://github.com/AgentEra/Agently"]},{"persona":"inversora","persona_name":"La Inversora","ai_generated":true,"verdict":"A small vendor, a permissive licence, no price and no hosted anything, so the survival question is entirely about one team's continued interest.","scores":{"reliability":5,"usefulness":5,"cost":8,"longevity":5},"evidence":["https://github.com/AgentEra/Agently","https://agently.tech"]},{"persona":"jefa","persona_name":"La Jefa","ai_generated":true,"verdict":"Free for sixty engineers and invisible to my operations: there is no unattended run mode, so nothing here is scheduled, monitored or attributable.","scores":{"reliability":4,"usefulness":5,"cost":8,"longevity":5},"evidence":["https://agently.tech/docs","https://github.com/AgentEra/Agently"]},{"persona":"hacker","persona_name":"El Hacker","ai_generated":true,"verdict":"Apache-2.0, one pip install, MCP servers are ordinary actions, and the README documents pointing the provider at a local Ollama endpoint on 127.0.0.1.","scores":{"reliability":8,"usefulness":7,"cost":9,"longevity":6},"evidence":["https://github.com/AgentEra/Agently/blob/main/docs/en/actions/mcp.md","https://github.com/AgentEra/Agently"]},{"persona":"comediante","persona_name":"El Comediante","ai_generated":true,"verdict":"The built-in actions include shell, Python, Node and SQLite, so the agent can now break four things without leaving the process.","scores":null,"evidence":[]}]}