BMAD-METHOD/.patch/821/subagentic/claude-subagents/AGENTS.md

Project Agents

This file provides guidance and memory for your coding CLI.

Opencode subagents and Tasks (OpenCode)

OpenCode reads AGENTS.md during initialization and uses it as part of its system prompt for the session.

How To Use With Claude

• Copy/paste claude-subagents subfolders in this project to ~/.claude and Claude will read and access agents from ~/.claude/agents and tasks from ~/.claude/tasks,
• You can access agents using "@ux-expert", or you can reference a role naturally, e.g., "As ux-expert, implement ..." or use commands defined in your tasks.

Note

• Orchestrators run as mode: primary; other agents as all.
• All agents have tools enabled: write, edit, bash.

Agents

Directory

Title	ID	When To Use
1-Create PRD	1-create-prd	1. Define Scope: use to clearly outlining what needs to be built with a Product Requirement Document (PRD)
2-Generate Tasks	2-generate-tasks	2. Detailed Planning: use to break down the PRD into a granular, actionable task list
3-Process Task List	3-process-task-list	3. Iterative Implementation: use to guide the AI to tackle one task at a time, allowing you to review and approve each change
UX Expert	ux-expert	Use for UI/UX design, wireframes, prototypes, front-end specifications, and user experience optimization
Scrum Master	scrum-master	Use for story creation, epic management, retrospectives in party-mode, and agile process guidance
Test Architect & Quality Advisor	qa-test-architect	Use for comprehensive test architecture review, quality gate decisions, and code improvement. Provides thorough analysis including requirements traceability, risk assessment, and test strategy. Advisory only - teams choose their quality bar.
Product Owner	product-owner	Use for backlog management, story refinement, acceptance criteria, sprint planning, and prioritization decisions
Product Manager	product-manager	Use for creating PRDs, product strategy, feature prioritization, roadmap planning, and stakeholder communication
Full Stack Developer	full-stack-dev	Use for code implementation, debugging, refactoring, and development best practices
Master Orchestrator	orchestrator	Use for workflow coordination, multi-agent tasks, role switching guidance, and when unsure which specialist to consult
Master Task Executor	master	Use when you need comprehensive expertise across all domains, running 1 off tasks that do not require a persona, or just wanting to use the same agent for many things.
Architect	holistic-architect	Use for system design, architecture documents, technology selection, API design, and infrastructure planning
Business Analyst	business-analyst	Use for market research, brainstorming, competitive analysis, creating project briefs, initial project discovery, and documenting existing projects (brownfield)

1-Create PRD (id: 1-create-prd)

Source: .agents/ux-expert.md

• When to use: Define Scope: use to clearly outlining what needs to be built with a Product Requirement Document (PRD) optimization
• How to activate: Mention "create prd, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

2-Generate Tasks (id: 2-generate-tasks)

Source: .agents/ux-expert.md

• When to use: 2. Detailed Planning: use to break down the PRD into a granular, actionable task list
• How to activate: Mention "generate tasks, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

3-Process Task List (id: 3-process-task-list)

Source: .agents/ux-expert.md

• When to use: 3. Iterative Implementation: use to guide the AI to tackle one task at a time, allowing you to review and approve each change
• How to activate: Mention "process task list, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

UX Expert (id: ux-expert)

Source: .agents/ux-expert.md

• When to use: Use for UI/UX design, wireframes, prototypes, front-end specifications, and user experience optimization
• How to activate: Mention "As ux-expert, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

Scrum Master (id: scrum-master)

Source: .agents/scrum-master.md

• When to use: Use for story creation, epic management, retrospectives in party-mode, and agile process guidance
• How to activate: Mention "As sm, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

Test Architect & Quality Advisor (id: qa-test-architect)

Source: .agents/qa-test-architect.md

• When to use: Use for comprehensive test architecture review, quality gate decisions, and code improvement. Provides thorough analysis including requirements traceability, risk assessment, and test strategy. Advisory only - teams choose their quality bar.
• How to activate: Mention "As qa, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

Product Owner (id: product-owner)

Source: .agents/product-owner.md

• When to use: Use for backlog management, story refinement, acceptance criteria, sprint planning, and prioritization decisions
• How to activate: Mention "As po, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

Product Manager (id: product-manager)

Source: .agents/product-manager.md

• When to use: Use for creating PRDs, product strategy, feature prioritization, roadmap planning, and stakeholder communication
• How to activate: Mention "As pm, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

Full Stack Developer (id: full-stack-dev)

Source: .agents/full-stack-dev.md

• When to use: Use for code implementation, debugging, refactoring, and development best practices
• How to activate: Mention "As dev, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

Master Orchestrator (id: orchestrator)

Source: .agents/orchestrator.md

• When to use: Use for workflow coordination, multi-agent tasks, role switching guidance, and when unsure which specialist to consult
• How to activate: Mention "As orchestrator, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

Master Task Executor (id: master)

Source: .agents/master.md

• When to use: Use when you need comprehensive expertise across all domains, running 1 off tasks that do not require a persona, or just wanting to use the same agent for many things.
• How to activate: Mention "As master, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

Architect (id: holistic-architect)

Source: .agents/holistic-architect.md

• When to use: Use for system design, architecture documents, technology selection, API design, and infrastructure planning
• How to activate: Mention "As architect, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

Business Analyst (id: business-analyst)

Source: .agents/business-analyst.md

• When to use: Use for market research, brainstorming, competitive analysis, creating project briefs, initial project discovery, and documenting existing projects (brownfield)
• How to activate: Mention "As analyst, ..." to get role-aligned behavior
• Full definition: open the source file above (content not embedded)

Tasks

These are reusable task briefs; use the paths to open them as needed.

Task: validate-next-story

Source: .tasks/validate-next-story.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: trace-requirements

Source: .tasks/trace-requirements.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: test-design

Source: .tasks/test-design.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: shard-doc

Source: .tasks/shard-doc.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: risk-profile

Source: .tasks/risk-profile.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: review-story

Source: .tasks/review-story.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: qa-gate

Source: .tasks/qa-gate.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: nfr-assess

Source: .tasks/nfr-assess.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: index-docs

Source: .tasks/index-docs.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: generate-ai-frontend-prompt

Source: .tasks/generate-ai-frontend-prompt.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: facilitate-brainstorming-session

Source: .tasks/facilitate-brainstorming-session.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: execute-checklist

Source: .tasks/execute-checklist.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: document-project

Source: .tasks/document-project.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: create-next-story

Source: .tasks/create-next-story.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: create-doc

Source: .tasks/create-doc.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: create-deep-research-prompt

Source: .tasks/create-deep-research-prompt.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: create-brownfield-story

Source: .tasks/create-brownfield-story.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: correct-course

Source: .tasks/correct-course.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: brownfield-create-story

Source: .tasks/brownfield-create-story.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: brownfield-create-epic

Source: .tasks/brownfield-create-epic.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: apply-qa-fixes

Source: .tasks/apply-qa-fixes.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)

Task: advanced-elicitation

Source: .tasks/advanced-elicitation.md

• How to use: Reference the task in your prompt or execute via your configured commands.
• Full brief: open the source file above (content not embedded)