Brain

Plugin brain in ansatzX/CyberBrain. Commands read from the repo's manifest — 1 month ago.

Skills (8)

• agentic-search

  Use when a task involves web search, current information, source verification, citations, or entity/person/paper/project disambiguation.

• codex-compatible

  Use for Codex exec_command permissions, sandbox paths, and host-native external CLI delegation when the task actually needs them.

• epistemic-systems-audit

  Use when reading papers, evaluating AI4S systems, reviewing scientific claims, judging benchmark or dataset results, checking claims about understanding/discovery/reasoning/autonomy, or repairing inflated scientific claims.

• state-machine

  Track meaningful project changes and verification, using an internal check for read-only tasks and a local ledger or state DAG when recovery and ordering matter.

• tame-dev-workflows

  Use when installing, reinstalling, auditing, or preparing Brain-managed Superpowers, Spec Kit, and OpenSpec workflow arsenals under a project-local `.brain/vendor` directory so Brain can inspect and route them without exposing their raw Codex skills/prompts at the project root.

• think-before-you-calculate

  Frame scientific calculations, simulations, experiments and benchmark interpretation so the measured proxy supports the intended claim. Ordinary builds, tests, scripts and agent calls do not trigger this audit by themselves.

• using-ansatz-brain

  Route Brain's object/evidence and state checks at conversation start, with a lightweight internal pass for ordinary work and detailed skills only when needed.

• whole-object-responsibility

  Trace the object, evidence, control, failure and ownership of a system; use a compact internal check ordinarily and a detailed chain for explicit system audits.

Similar plugins