AI Agent Workflow Collection

Push the LLM to reconsider, refine, and improve its recent output. Use when user asks for deeper critique or mentions a known deeper critique method, e.g. socratic, first principles, pre-mortem, red team.

Strategic business analyst and requirements expert. Use when the user asks to talk to Mary or requests the business analyst.

System architect and technical design leader. Use when the user asks to talk to Winston or requests the architect.

Senior software engineer for story execution and code implementation. Use when the user asks to talk to Amelia or requests the developer agent.

Product manager for PRD creation and requirements discovery. Use when the user asks to talk to John or requests the product manager.

Technical documentation specialist and knowledge curator. Use when the user asks to talk to Paige or requests the tech writer.

UX designer and UI specialist. Use when the user asks to talk to Sally or requests the UX designer.

Facilitate interactive brainstorming sessions using diverse creative techniques and ideation methods. Use when the user says help me brainstorm or help me ideate.

LLM-assisted human-in-the-loop review. Make sense of a change, focus attention where it matters, test.

Course correction navigator for mid-sprint scope changes. Loads sprint state from the issue tracker, runs a structured impact assessment (affected epics, stories to modify/create/cancel, architecture changes, risk evaluation), proposes corrective actions, and executes approved changes in the tracker. Use when 'correct course', 'correction trajectoire', 'course correction', 'changement scope', 'scope change', 'pivot sprint' is mentioned.

Collaborative architecture solution design for AI agent implementation consistency. Guides through system context analysis, technology decisions, pattern selection, conflict resolution, structure definition, and validation. Produces an architecture document that ensures consistent code generation across all AI agents. Use when 'create architecture', 'solution design', 'technical architecture', 'créer architecture', 'design technique', 'architecture solution', 'décisions techniques' is mentioned.

Structured PRD creation through collaborative product discovery. Captures vision, success criteria, user journeys, domain requirements, and functional/non-functional specifications through step-by-step facilitation. Use when 'create prd', 'new prd', 'product requirements', 'créer prd', 'nouveau prd', 'exigences produit' is mentioned.

Unified story creation and enrichment engine (story-spec v2). Produces implementation-ready stories with real-data confrontation (provider/DB/cloud), external research (docs/RFC/gotchas), structured NFRs (7 categories), binary security gate, observability requirements, deployment chain audit, impact analysis, BACs in Given/When/Then, TACs in EARS notation, INVEST self-check, out-of-scope register, risks/assumptions register, boundaries triple, and multi-validator review. Discovery mode creates a new tracker issue from conversational spec engineering; Enrichment mode loads full context (PRD, Architecture, UX, completed stories, git history) and enriches an existing Backlog issue. Use when 'create story', 'enrich story', 'spec story', 'write spec', 'nouvelle story', 'enrichir story', 'preparer story', 'creer story' is mentioned.

UX design specification through collaborative discovery. Captures core experience, emotional design goals, design system, visual foundations, user journeys, component strategy, and responsive/accessibility patterns. Use when 'create ux', 'ux design', 'design ux', 'ux specifications', 'créer ux', 'spécifications ux', 'plan ux', 'design interface' is mentioned.

Authors and updates customization overrides for installed BMad skills. Use when the user says 'customize bmad', 'override a skill', 'change agent behavior', or 'customize a workflow'.

Lossless LLM-optimized compression of source documents. Use when the user requests to 'distill documents' or 'create a distillate'.

Document brownfield projects for AI context by scanning source code and generating comprehensive documentation. Detects project type, scans codebase at configurable depth (quick/deep/exhaustive), and produces structured documentation for LLM consumption. Supports resume from previous scans. Use when 'document project', 'generate project docs', 'scan codebase', 'documenter le projet', 'scanner le code', 'documentation projet' is mentioned.

Conduct domain and industry research.

Edit and improve existing PRDs through validation-guided refinement. Loads PRD, detects format (BMAD/legacy), reviews against quality standards, and applies targeted edits with user guidance. Use when 'edit prd', 'update prd', 'improve prd', 'modifier prd', 'ameliorer prd', 'corriger prd', 'refine prd' is mentioned.

Clinical copy-editor that reviews text for communication issues.

Structural editor that proposes cuts, reorganization, and simplification while preserving comprehension.

Analyzes current state and user query to answer BMad questions or recommend the next skill(s) to use. Use when user asks for help, bmad help, what to do next, or what to start with in BMad.

Generates or updates an index.md to reference all docs in the folder.

Conduct market research on competition and customers.