Overview

Add a "Resume DT Coaching" handoff entry to the task-researcher agent definition, enabling users to return to DT coaching after completing RPI research tasks. This creates the reverse path of the DT→RPI handoff: when task-researcher completes research triggered by a Problem Space exit, the user can return to the DT coach with research findings integrated.

Target File

Modify .github/agents/task-researcher.agent.md. The agent's YAML frontmatter contains a handoffs: array with one existing entry (task-planner). The modification adds a second handoff entry.

Required Changes

New Handoff Entry

Add to task-researcher's handoff declarations:

Frontmatter diff — add to the handoffs: array in .github/agents/task-researcher.agent.md:

- agent: dt-coach
  description: 'Resume DT Coaching — return to DT coaching with research findings integrated'

Handoff Behavior

When the "Resume DT Coaching" handoff is triggered:

1. Task-researcher summarizes research findings
2. Research output location is noted for the DT coach to reference
3. Handoff to dt-resume-coaching prompt (or directly to dt-coach agent at .github/agents/dt-coach.agent.md, defined in feat(agents): Create dt-coach.agent.md #579)
4. DT coach reads updated coaching state and research findings
5. Coach re-establishes coaching context and continues from where the user left off — potentially with new evidence that changes the DT trajectory

Research Integration

The DT coach receives:

• Original coaching state (preserved during RPI handoff)
• New research findings from task-researcher
• Any insights or recommendations generated during research

The coach can:

• Incorporate findings into the current method's artifacts
• Suggest revisiting earlier methods if findings invalidate previous synthesis
• Recommend proceeding to the next method with enriched data

Implementation Scope

This is a minimal modification to the task-researcher agent:

• Add one handoff entry to the agent's handoff table
• Add brief description of the return path behavior
• No structural changes to the task-researcher agent

RPI Pipeline Workflow

1. task-researcher: Locate the current task-researcher agent definition file. Identify the handoff section and understand existing handoff patterns.
2. task-planner: Plan the modification — one new handoff entry, brief behavior description.
3. task-implementor: Add the handoff entry following existing patterns in the agent definition.
4. task-reviewer: Validate handoff entry consistency with existing patterns, verify dt-coach target, prompt-builder compliance.

Success Criteria

• "Resume DT Coaching" handoff added to task-researcher agent definition
• Handoff targets dt-coach or dt-resume-coaching prompt
• Handoff behavior describes research summary and context restoration
• Modification follows existing handoff patterns in the agent file
• 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