Prp Research Team

1. Strip --orchestration "..." from input → store as ORCHESTRATION
2. Remaining text = research question
3. If empty after parsing → STOP with usage error

Identify Scope Signals

GATE: If the research question is too vague to decompose → STOP and ASK for clarification.

Phase 2: CLASSIFY — Domain & Complexity

Determine Research Domain

Assess Complexity

If ORCHESTRATION is set, adjust team composition, domain weighting, and expertise requirements accordingly.

CHECKPOINT: Domain identified, complexity assessed, orchestration applied.

Phase 3: DECOMPOSE — Sub-Questions

3.1 Break Down Research Question

Decompose into 3-7 independently investigable sub-questions.

Rules:

1. Each sub-question must be answerable by a single researcher
2. Sub-questions must cover the full scope of the original question
3. Identify PARALLEL vs. DEPENDENT sub-questions
4. Tag each with its primary domain

3.2 Map Dependencies

SQ-1 (foundational) ──┬──► SQ-2 (parallel)
                       ├──► SQ-3 (parallel)
                       └──► SQ-4 (parallel)
                                    │
                                    ▼
                              SQ-5 (synthesis, depends on SQ-2,3,4)

Dependency types: NONE (start immediately), BLOCKED_BY (must wait), INFORMS (benefits from but doesn't require).

3.3 Validate Coverage

Verify sub-questions collectively cover the full scope, don't overlap significantly, and include at least one synthesis sub-question.

CHECKPOINT: 3-7 sub-questions defined, dependencies mapped, coverage verified.

Phase 4: COMPOSE — Design Team Roles

Design researcher profiles. Load the templates and model selection guide:

${CLAUDE_SKILL_DIR}/references/team-templates.md

Each researcher must have: Name, Focus, Sub-questions, Model, Spawn prompt, Output format, Completion criteria.

Every spawn prompt must be self-contained — the agent must execute with ONLY that prompt, no external context.

CHECKPOINT: All researchers defined with complete profiles, spawn prompts self-contained, team covers all sub-questions.

Phase 5: PLAN — Research Tasks

5.1 Create Task List

For each task define: ID (RT-{N}), Title, Assignee, Type (RESEARCH/ANALYSIS/SYNTHESIS/REVIEW), Dependencies, Description, Acceptance criteria, Estimated effort (LOW/MEDIUM/HIGH).

5.2 Task Ordering

1. Wave 1: All tasks with no dependencies (parallel)
2. Wave 2: Tasks depending on Wave 1 outputs
3. Wave 3: Synthesis and integration tasks
4. Final: Review and quality assurance

5.3 Cross-Cutting Concerns

Define shared standards: citation format, confidence level tagging (HIGH/MEDIUM/LOW with rationale), contradiction handling, scope boundary enforcement.

CHECKPOINT: Tasks defined with valid DAG dependencies, parallel tasks identified, synthesis task exists.

Phase 6: GENERATE — Write Research Plan

6.1 Setup

mkdir -p .claude/PRPs/research-plans

6.2 Write State Sentinel

Write the output path to .claude/prp-research-team.state (one line, just the file path) for stop hook integration.

6.3 Write Research Plan

Load the full plan template from:

${CLAUDE_SKILL_DIR}/references/team-templates.md

Write to: .claude/PRPs/research-plans/{topic-slug}.research-plan.md

GATE: Do NOT proceed until the plan has ALL 6 required sections:

1. ## Research Question
2. ## Research Question Decomposition
3. ## Team Composition
4. ## Research Tasks
5. ## Team Orchestration Guide
6. ## Acceptance Criteria

CHECKPOINT: Plan written with all sections, sentinel file set, no placeholders remain.

Phase 7: OUTPUT — Report to User

Load the user output template from:

${CLAUDE_SKILL_DIR}/references/team-templates.md

Display: file path, question, team composition table, plan overview (domain, complexity, task counts by wave), and execution instructions for both agent teams and manual sequential approaches.

CHECKPOINT: Summary displayed, team shown, execution instructions provided.

Success Criteria

• QUESTION_PARSED: Research question extracted and validated
• DOMAIN_CLASSIFIED: Primary and supporting domains identified
• DECOMPOSED: 3-7 independent sub-questions with dependency mapping
• TEAM_DESIGNED: Each researcher has name, focus, spawn prompt, output format, completion criteria
• TASKS_PLANNED: All tasks have IDs, assignees, dependencies, acceptance criteria
• PLAN_WRITTEN: Research plan file created with all required sections
• SENTINEL_SET: State file written for stop hook validation
• USER_INFORMED: Summary with execution instructions displayed