agentbrain

Agent Brain /brain-brief: after intake, research, and grill have enough signal.

Agent Brain /brain-build: an implementation plan has a selected task and validation method.

Agent Brain /brain-design: a product brief needs ux or interaction design before planning.

Agent Brain /brain-eval: changing Agent Brain behavior or checking quality.

Agent Brain /brain-grill: the idea, brief, design, or plan has unresolved assumptions.

Agent Brain /brain-learn: after repeated success/failure, a tricky fix, or a shipped workflow.

Agent Brain /brain-plan: the brief/design is strong enough to implement.

Agent Brain /brain-research: the decision depends on external evidence or unfamiliar domain context.

Agent Brain /brain-review: after verification or before public/shipping decisions.

Agent Brain /brain-ship: a reviewed artifact is ready for release or publication.

Agent Brain /brain-should-this-exist: before planning any new product, feature, workflow, or automation.

Agent Brain /brain-start: a user starts from a vague request, idea, task, or product ambition.

Agent Brain /brain-verify: after build or when evaluating an existing artifact.

Agent Brain /brain-wiki: ingesting sources or updating durable project knowledge.

Use when deciding what project context should be remembered, retrieved, updated, or deliberately forgotten.

Use when verification depends on agent runtime phases, queued input, tool-call lifecycle, save points, retry, abort, compaction, or branch behavior.

Use when the user needs a standup-ready summary of recent project activity from local evidence.

Use when an output artifact, template, schema, example, or handoff field must stay aligned across commands, skills, docs, and validators.

Use when an agent runtime must choose or verify a /brain-* command route from a user request, handoff, adapter transcript, or stale command catalog.

Use when checking Agent Brain inside a real agent runtime or adapter before trusting harness usability.

Use when a command, adapter, or runtime smoke depends on proving an agent runtime's capabilities before trusting command routing, writes, shell access, or validation claims.

Use when agent-produced work needs a safety, evidence, and reliability check before handoff, merge, or trust.

Use when local validation and remote workflow status must be reconciled before trust, merge, or shipment.

Use when a design needs pressure-testing for UX, information architecture, states, and accessibility.

Use when project vocabulary is fuzzy, overloaded, disputed, or needed before naming docs, schemas, commands, skills, code, plans, or memory.

Use when an engineering plan needs pressure-testing for architecture, complexity, data, risk, and operability.

Use when sources need to be converted into a claim ledger with confidence levels.

Use when raw intent must be captured and routed to the right next state.

Use when a launch needs a go/no-go decision with rollback and monitoring evidence.

Use when repeated wins or failures should become durable memory, wiki, or skill updates.

Use when a product idea needs pressure-testing against alternatives, category, urgency, and distribution.

Use when broad work needs to be broken into small verifiable slices.

Use when a product or feature idea needs pressure-testing against problem, user, pain, success, and timing.

Use when behavior needs concrete proof before review, merge, or shipment.

Use when uncertainty remains and the minimum useful sequence of questions should be asked.

Use when source-backed project knowledge needs to be created, refreshed, or corrected.