Overview

Create dt-rpi-research-context.instructions.md — a DT-aware instruction file that auto-loads when task-researcher operates on DT artifacts. This frames research activities around DT methods, stakeholder needs, and empathy-driven inquiry rather than pure technical research patterns.

When task-researcher receives a handoff from the DT coach (via Problem Space exit), this instruction file helps the researcher understand that:

• The research question comes from a DT process, not a technical spec
• Stakeholder perspectives drive the research framing
• Findings should feed back into the DT framework (validated/assumed/unknown markers)
• The researcher may need to return findings to the DT coach rather than proceeding to planning

Directory context: The .copilot-tracking/dt/ path in applyTo is created per-project by the coach agent (#579 from Phase 2). See #583 for directory structure and handoff contract details.

Target File

.github/instructions/dt-rpi-research-context.instructions.md

Frontmatter

---
description: 'DT-aware task-researcher context — frames research around DT methods, stakeholder needs, and empathy-driven inquiry'
applyTo: '**/.copilot-tracking/dt/**'
---

Required Content

Research Framing Adjustments

When operating in DT context, task-researcher adjusts its approach:

DT-Specific Research Patterns

• Stakeholder-driven queries: When the research question mentions stakeholders, investigate from multiple stakeholder perspectives
• Assumption validation: When the handoff artifact flags "assumed" items, prioritize validating or invalidating those assumptions
• Gap investigation: When the handoff artifact flags "unknown" items, these become primary research targets
• Empathy data synthesis: Process qualitative research data (interview themes, observation patterns) alongside quantitative data
• Industry context integration: If an industry context template was used during coaching, reference it for domain-specific terminology and constraints

Output Format Additions

Task-researcher adds DT-specific sections to research output when operating in DT context:

• Assumption Validation Results: For each "assumed" item from the handoff, provide evidence status
• Gap Resolution: For each "unknown" item, provide findings or recommend continued investigation
• Stakeholder Impact Assessment: How research findings affect different stakeholder groups
• DT Coach Return Recommendation: Whether findings warrant returning to DT coaching before proceeding to planning

Return Path Awareness

Task-researcher should recommend returning to DT coaching when:

• Research reveals that the problem statement needs significant revision
• New stakeholders are discovered that weren't in the original map
• Fundamental assumptions are invalidated
• Conflicting evidence suggests the synthesis (Method 3) needs rework

Token Budget

Target: ~500-700 tokens (ambient tier — auto-loads on all DT artifact paths)

RPI Pipeline Workflow

1. task-researcher: Review existing task-researcher agent definition and behavioral patterns. Understand how applyTo globs modify researcher behavior. Gather DT research framing from cumulative research.
2. task-planner: Plan the file — research framing adjustments, DT-specific patterns, output additions, return path triggers.
3. task-implementor: Author following prompt-builder standards. Adjustments should augment (not replace) standard researcher behavior.
4. task-reviewer: Validate against DT review context (feat(instructions): Create dt-rpi-review-context.instructions.md #593), confirm research framing adjustments, prompt-builder compliance.

Success Criteria

• Instruction file created at .github/instructions/dt-rpi-research-context.instructions.md
• Frontmatter has applyTo: '**/.copilot-tracking/dt/**' glob
• Research framing adjustments defined (stakeholder-driven, multi-perspective, quality-marked)
• DT-specific research patterns cover assumption validation, gap investigation, empathy data synthesis
• Output format additions include assumption validation results and stakeholder impact
• Return path triggers defined (problem statement revision, new stakeholders, invalidated assumptions)
• Token count within ~500-700 target
• Passes task-reviewer validation against prompt-builder standards
• Each prompt, instructions, or agent file registered in collections/design-thinking.collection.yml with path and kind fields
• Each prompt, instructions, or agent file registered in collections/hve-core-all.collection.yml with path and kind fields
• npm run plugin:generate succeeds after collection manifest updates