kanbanzai

Use at the start of every agent session, even if the task seems obvious and you think you already know what to do. Activates when the agent has just opened a repository, does not know what to do, needs to orient itself, or is beginning any new session. Also activates for "where do I start?", "what should I work on?", "what is the current state?". Skipping orientation leads to wasted effort and missed context.

expert: "Structured task execution procedure for individual implementation

expert: "Multi-agent development orchestration with conflict-aware parallel

expert: "Review orchestration with adaptive specialist dispatch,

Use at the start of every agent session, even if the task seems obvious and you think you already know what to do. Activates when the agent has just opened a repository, does not know what to do, needs to orient itself, or is beginning any new session. Also activates for "where do I start?", "what should I work on?", "what is the current state?". Skipping orientation leads to wasted effort and missed context.

expert: "Structured task execution procedure for individual implementation

expert: "Multi-agent development orchestration with conflict-aware parallel

expert: "Review orchestration with adaptive specialist dispatch,

expert: "Execute a 10-item Definition of Done checklist with concrete

Use when dispatching implementation tasks to sub-agents, committing changes, finishing tasks, or contributing knowledge. Activates when there is a task to implement, a commit to make, or an entity to complete. Also activates for entity creation, naming, and data entry decisions.

Use when creating, editing, registering, or approving documents in a Kanbanzai-managed project. Activates for document types, placement, registration with doc action register, approval workflow, content drift, doc action refresh, or any question about document status, ownership, or lifecycle. Use whenever any markdown file is created or modified in a configured document root — even if the edit seems minor, registration and refresh rules still apply.

expert: "Performs a spec-free, whole-codebase quality audit by running static

expert: "Feature-to-task decomposition procedure using the decompose tool

expert: "Prompt and skill authoring following attention-optimised section

expert: "Multi-dimension code review producing classified findings

expert: "Design document authoring producing a structured 4-section document

expert: "Implementation plan authoring producing a structured 5-section

expert: "Research report authoring producing a structured investigation document

expert: "Specification authoring producing a gate-checkable 5-section specification

expert: "Prompt and skill authoring following attention-optimised section

Use this skill when conducting a plan review after all features in a plan are complete. Covers scope verification, feature completion checks, spec conformance, documentation currency, and retrospective contribution.

Use when starting a new body of work, scoping a feature or plan, deciding what to build next, or determining whether something is one feature or many. Also activates for planning, scoping, and ambition questions: "is this too big?", "should we split this?", "how ambitious should we be?", "what would a great team do here?", or any question about prioritisation, roadmap, or scope. Use even when the user doesn't explicitly say "planning" — any discussion about what to build next is a planning conversation. Use even for seemingly small changes — unplanned work accumulates into incoherent systems.

expert: "Documentation QA producing classified findings for factual accuracy,

expert: "Sentence-level copy editing enforcing active voice, verb clarity,

expert: "Structural and developmental editing verifying inverted-pyramid

expert: "Sequential pipeline coordination dispatching five editorial

expert: "Batch-level conformance review verifying feature delivery status,

expert: "AI-artifact detection and removal targeting banned vocabulary,

expert: "Documentation maintenance and update skill producing consistent,

expert: "First-draft technical documentation following inverted-pyramid

expert: "Skill authoring producing a research-backed SKILL.md with vocabulary

Use when deciding what workflow stage work belongs to, whether to proceed to the next stage, whether a state transition is valid, or when to stop and ask the human. Activates for lifecycle, stage gates, entity status, approval requirements, or uncertainty about proceeding. Use even when the agent is confident — workflow gates exist precisely for the cases where confidence is misplaced.

Use when creating, editing, registering, or approving documents in a Kanbanzai-managed project. Activates for document types, placement, registration with doc action register, approval workflow, content drift, doc action refresh, or any question about document status, ownership, or lifecycle. Use whenever any markdown file is created or modified in a configured document root — even if the edit seems minor, registration and refresh rules still apply.

expert: "Implementation plan authoring producing a structured 5-section

expert: "Research report authoring producing a structured investigation document

expert: "Specification authoring producing a gate-checkable 5-section specification

Guides you through implementing a single task — read what's required, build it, test it, verify it matches the spec

Use at the start of every session to orient yourself, find what to work on, and check the current project state

Use when deciding workflow stage transitions, stage gates, entity lifecycle rules, or whether to stop and ask the human

Multi-agent development orchestration — dispatch parallel tasks, monitor progress, handle failures, and close out the feature lifecycle