{"slug":"claudectl","name":"claudectl","vendor":"mercurialsolo","tagline":"6 MB Rust orchestrator for a swarm of Claude Code agents, backed by a local- LLM brain that learns from your own sessions","url":"https://agentboards.org/agent/claudectl","website":"https://mercurialsolo.github.io/claudectl/","docs":null,"repo":"https://github.com/mercurialsolo/claudectl","category":"harness","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":["Claude Code","local models"],"bring_your_own_model":true,"local_models":true},"protocols":{"mcp_client":false,"mcp_server":false,"openapi":false},"capabilities":{"terminal_exec":true,"browser":false,"multi_file_edit":true,"git_ops":true,"docker_sandbox":false,"multi_agent":true,"headless_ci":false},"context_window":null,"platforms":["macos","linux"],"install":[{"label":"Cargo","command":"cargo install claudectl"},{"label":"Homebrew","command":"brew install mercurialsolo/tap/claudectl"}],"benchmarks":[],"tags":["open-source","rust","harness","claude-code","swarm","local-models"],"github_stars":201,"metrics":{"stars":201,"pushed_at":"2026-07-10T04:00:56Z","archived":false,"open_issues":63,"latest_version":"v0.64.0","latest_version_at":"2026-07-04T03:42:11Z","latest_version_url":"https://github.com/mercurialsolo/claudectl/releases/tag/v0.64.0","version_source":"github","checked_at":"2026-10-02T05:04:59.144Z"},"sources":{"overview":"https://github.com/mercurialsolo/claudectl","capabilities":"https://github.com/mercurialsolo/claudectl","models":"https://github.com/mercurialsolo/claudectl","license":"https://github.com/mercurialsolo/claudectl","install":"https://crates.io/crates/claudectl","website":"https://mercurialsolo.github.io/claudectl/"},"verified_at":"2026-09-04","adoption":{"score":0,"signals":{"github_stars":201,"hn_mentions_1y":3},"measured_at":"2026-09-21"},"scores":{"panel":4.9,"adoption":0,"spread":3,"facts":6.7,"board":4.4,"dimensions":{"reliability":4.5,"usefulness":4.7,"cost":6.5,"longevity":4},"per_persona":{"amigo":5.3,"critico":4.8,"profesor":4.5,"inversora":4.5,"jefa":3.8,"hacker":6.8},"community":null},"rank":null,"category_rank":193,"panel_reviews":[{"persona":"juez","persona_name":"El Juez","ai_generated":true,"verdict":"El Profesor and El Hacker both stop at the same component and reach opposite conclusions about whether learning on your machine is a feature.","scores":null,"evidence":[]},{"persona":"amigo","persona_name":"El Amigo","ai_generated":true,"verdict":"Pick it if you already run several sessions at once and want them coordinated; pick a single agent if one window is still enough work for one day.","scores":{"reliability":5,"usefulness":5,"cost":7,"longevity":4},"evidence":["https://github.com/mercurialsolo/claudectl"]},{"persona":"critico","persona_name":"El Crítico","ai_generated":true,"verdict":"A learning component absorbs your sessions and shares what it concludes across every agent, and nothing documented says how a wrong conclusion is removed.","scores":{"reliability":4,"usefulness":5,"cost":6,"longevity":4},"evidence":["https://github.com/mercurialsolo/claudectl"]},{"persona":"profesor","persona_name":"El Profesor","ai_generated":true,"verdict":"The impact scorecard is produced by the tool measuring itself, using numbers it accumulated, with no stated method and no external comparison.","scores":{"reliability":4,"usefulness":4,"cost":6,"longevity":4},"evidence":["https://github.com/mercurialsolo/claudectl"]},{"persona":"inversora","persona_name":"La Inversora","ai_generated":true,"verdict":"197 stars, a solo author publishing through a package registry and a personal tap, and no company anywhere in the picture to fund a second year.","scores":{"reliability":4,"usefulness":4,"cost":7,"longevity":3},"evidence":["https://github.com/mercurialsolo/claudectl"]},{"persona":"jefa","persona_name":"La Jefa","ai_generated":true,"verdict":"Free across sixty desks and sixty provider subscriptions underneath it, with nothing that runs unattended and no administrative surface of any kind.","scores":{"reliability":3,"usefulness":4,"cost":5,"longevity":3},"evidence":["https://github.com/mercurialsolo/claudectl"]},{"persona":"hacker","persona_name":"El Hacker","ai_generated":true,"verdict":"MIT, one static Rust binary I can drop anywhere, and local weights are a supported destination rather than a footnote about future work.","scores":{"reliability":7,"usefulness":6,"cost":8,"longevity":6},"evidence":["https://github.com/mercurialsolo/claudectl"]},{"persona":"comediante","persona_name":"El Comediante","ai_generated":true,"verdict":"Six megabytes and it starts in under fifty milliseconds, then waits nine seconds for a model to finish thinking.","scores":null,"evidence":[]}]}