Agent Claude CodeOpenCode
Verify BioMaster step outputs, maintain CHECKREPORT.yaml and PROJECTLOG.yaml, and emit structured check results.
Agent Claude CodeOpenCode
Verify BioMaster step outputs, maintain CHECKREPORT.yaml and PROJECTLOG.yaml, and emit structured check results.
Agent Claude CodeOpenCode
Inspect BioMaster script execution status and propose bounded retry advice.
Agent Claude CodeOpenCode
Execute BioMaster plan steps using only the exec skills referenced by PLAN.yaml.
Agent Claude CodeOpenCode
Primary BioMaster agent. Collects goals, selects/reuses a project folder, routes through SQLite, generates PLAN.yaml, and coordinates workers.
Agent Claude CodeOpenCode
Skill curator. Evaluates raw material, and (re)divides it into a curated macro/micro + structured-steps BioSkill via the bioskills-author tool.
Agent Claude CodeOpenCode
Skill generator. Synthesizes a brand-new curated BioSkill from a paper / URL / GitHub repo plus a required skill-focus description, using webfetch to read source material and the bioskills-author tool to write.
Agent Claude CodeOpenCode
Skill router. Decides which BioSkills layer (macro/micro) each sub-agent sees and fetches it via the bioskills tool.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: