agent-team

Design, create, evolve, or audit Codex-native harnesses for Agent Team projects. Builds specialized .codex/agents, .agents/skills, an orchestrator skill, AGENTS.md pointers, validation, and an evolution loop. Use for 'set up codex harness', 'create codex automation', 'build specialist agents', harness updates, harness audits, sync, status, and follow-up refinement. Generated harnesses default orchestrated runtime execution to Agent Team runtime skills backed by daemonless `agent-team` state.

agent-team: Complete an assigned task with evidence and artifact path.

Designer persona for agent-team design work. Routes between deep design interview (brief) and artifact production (spec). Use for 'design this', 'design discovery', 'design interview', 'produce design spec', 'visual/UI/icon/character/environment/logo/design-system design'. Skip for backend architecture (use persona-agent-team-planner), implementation code, code review, run execution, worker checkpoints, audits, or learning capture.

Planner persona for agent-team pre-execution work. Routes between terminology alignment, plan stress-testing, and backend architecture design before coding starts. Use for 'plan this', 'align terms', 'grill the plan', 'propose backend structure', or any pre-implementation discovery for code/architecture. For visual/UI/icon/character/environment/logo/design-system work use persona-agent-team-designer instead. Do not use for run execution, worker checkpoints, audits, learning capture, or direct coding.

Deep design interview that produces a design brief. Socratic 4-line probes lock the design's output description, tensions, assumptions, priorities, and failure modes. Output: design-brief.md. Skip for backend architecture, terminology cleanup, fuzzy planning, implementation, review, post-run learning, or rendered assets. Hand off to recipe-agent-team-design-spec for artifact production.

Generic spec producer for locked design briefs. Consumes design-brief.md, reads the Output description, consults pattern hints from references/, and produces artifacts under the brief's Output Path. Skip when no brief exists; run recipe-agent-team-design-interview first. Skip for backend architecture, code, review, or post-run learning.

Recipe: Stress-test a plan against repo code, docs, domain terms, and agent-team runtime constraints before execution. Use for 'grill the plan', 'challenge approach', 'refine planning', 'sharpen terminology', 'create acceptance criteria', or turning fuzzy ideas into durable tasks. Skip when plan is concrete with clear acceptance criteria and task contracts; go to architecture-design, coding, review, or compound-learning.

Recipe: Capture reusable learnings from completed agent-team runs, tasks, reviews, or fixes so future agents skip re-discovery. Use for 'compound learning', 'document what worked', 'capture a reusable pattern', 'summarize lessons', 'update solution docs', 'close the loop'. Do not use for in-progress runs, uncertain outcomes, trivial fixes, active implementation, active planning/design, or performing code review itself.

Design, create, evolve, or audit Gemini-native harnesses for Agent Team projects. Builds specialized .gemini/agents, .gemini/skills, an orchestrator skill, GEMINI.md pointers, validation, and an evolution loop. Use for 'set up gemini harness', 'create gemini automation', 'build specialist agents', Gemini harness updates, harness audits, sync, status, and follow-up refinement. Generated harnesses default orchestrated runtime execution to Agent Team runtime skills backed by daemonless `agent-team` state.

agent-team: Detect old in-progress or blocked tasks for daemonless operational recovery.

agent-team: Inspect append-only state events with run, entity, type, and version filters.

agent-team: Acknowledge a message after handling it.

agent-team: Read messages for one recipient agent.

agent-team: List messages in a run with optional audit filters.

agent-team: Send a compact run-scoped message to another agent.

agent-team: Cancel an open workflow run with a reason.

agent-team: Close a workflow run after all tasks are in a terminal state (done, failed, or cancelled).

agent-team: Create a daemonless workflow run.

agent-team: List workflow runs, optionally filtered by status.

agent-team: Show one workflow run and grouped task status counts.

agent-team: Show operational summary for one run, including blockers, unread messages, recent events, and close readiness.

agent-team: Export the machine-readable CLI command and error contract.

agent-team: Check unresolved inbox messages and incomplete dependencies before proceeding.

agent-team: Mark an assigned task blocked with an actionable reason.

agent-team: Cancel a non-terminal task with a reason.

agent-team: Create an assigned task inside a run.

agent-team: Mark assigned work failed with a reason and optional artifact.

agent-team: List tasks with optional run, agent, or status filters.

agent-team: Reassign a pending or blocked task to another agent.

agent-team: Reset blocked, in-progress, or failed work to pending.

agent-team: Show one task and its dependency IDs.

agent-team: Mark an assigned task in progress.

Recipe: Produce architecture design artifacts (modules, interfaces, task contracts, candidates) before coding. Use for 'design the architecture', 'propose a structure', 'evaluate candidates', 'write a technical design', 'break this into implementation tasks'. Do not use for vague goals, terminology alignment, implementation, code review, or post-run learning; use planning-grill, terminology-context, coding workflow, review workflow, or compound-learning instead.

Recipe: Read-only daemonless audit of agent-team runs, messages, events, stale tasks, unresolved inboxes, and CLI schema. Use for operational inspection, recovery assessment, runtime debugging, trace review, or contract export. Do not use to create runs, dispatch workers, mutate state, plan, design, align terms, compound learnings, or implement.

Service-level skill for agent-team message and inbox commands. Use for asynchronous daemonless communication, inbox checkpoint checks, acknowledgements, and compact contract-change notifications. Load agent-team-shared first.

Service-level skill for agent-team operational commands. Use for run hygiene, message inspection, event history, schema export, audits, and tooling. Load agent-team-shared first.

Service-level skill for agent-team run commands. Use when creating, listing, checking, or closing daemonless workflow runs. Load agent-team-shared first.

Shared runtime procedure for the agent-team CLI. Use for daemonless state, global flags, JSON input/output, state directory rules, error handling, and shell usage conventions. Load this before command-specific agent-team skills.

Service-level skill for agent-team sync checks. Use before workers complete tasks, when an orchestrator suspects drift, or when inbox, dependency, or state-version mismatches must be detected without a daemon. Load agent-team-shared first.

Service-level skill for agent-team task lifecycle commands. Use when creating, listing, showing, starting, completing, blocking, reassigning, retrying, cancelling, failing, or detecting stale assigned work. Load agent-team-shared first.