Overview

Create dt-handoff-solution-space.prompt.md — the Solution Space exit handoff prompt. When a user completes Methods 4-6 (Brainstorming → Concepts → Lo-Fi Prototypes) and is ready to transition, this prompt generates an RPI-ready artifact targeting either task-researcher (for feasibility deep-dive) or task-planner (for implementation planning). The generated output carries forward the selected concept with its D/F/V evaluation, prototype learnings, and constraint map.

Target File

.github/prompts/dt-handoff-solution-space.prompt.md

Frontmatter

---
description: 'Solution Space exit handoff — produces RPI-ready prompt from DT concept validation work'
mode: 'agent'
tools: ['read_file', 'create_file']
---

Required Content

Handoff Context

The Solution Space exit carries richer artifacts than the Problem Space exit:

• Selected concept(s): Full concept cards with D/F/V evaluation from Method 5
• Prototype learnings: What was validated, what was invalidated, what remains unknown from Method 6
• Constraint map: Technical, organizational, and market constraints identified during prototyping and evaluation
• Image prompts generated: Any M365 Copilot visualization prompts used (for continuity)
• Unresolved D/F/V gaps: Open questions per lens that need downstream resolution

Target Routing

The prompt determines the appropriate RPI entry point based on maturity:

• → task-researcher: When feasibility gaps remain — unanswered technical questions, unvalidated assumptions, missing market data. The generated prompt pre-loads the researcher with concept context and specific research questions.
• → task-planner: When the concept is sufficiently validated and ready for implementation planning. The generated prompt pre-loads the planner with concept specifications, constraint map, and prototype learnings.

Tiered Exit Schema

Follow the handoff contract tiered exit schema:

• Tier 1 (Guided): Early graduation — concept card and key learnings, may return to DT methods
• Tier 2 (Structured): Standard graduation — full D/F/V evaluation, prototype learnings, researcher or planner targeting
• Tier 3 (Comprehensive): Deep graduation — complete artifact set with constraint maps, image prompts, and explicit recommendations for RPI pipeline steps

The coach assesses readiness and suggests the appropriate tier. Users can choose to graduate at any tier.

Validated-vs-Unknown Markers

The generated artifact explicitly marks:

• [VALIDATED] — Claims backed by prototype testing or user feedback
• [ASSUMED] — Claims not yet tested but believed reasonable
• [UNKNOWN] — Open questions requiring downstream research or testing

These markers transfer into the RPI pipeline so task-researcher or task-planner knows what needs verification.

Required Steps

1. Step 1: Read Coaching State — Load coaching state to confirm Solution Space completion (Methods 4-6).
2. Step 2: Compile DT Artifacts — Read Method 4-6 output directories:
   • Method 4: brainstorming outputs, convergence decisions
   • Method 5: concept cards, D/F/V evaluation matrices, image prompts
   • Method 6: prototype specs, feedback summaries, constraint maps
3. Step 3: Readiness Assessment — Validate artifact completeness against the Solution Space exit schema from the handoff contract. Flag gaps and ask user whether to proceed or return to address them.
4. Step 4: Produce Handoff Artifact — Generate the exit-point artifact with:
   • Selected concept(s) with D/F/V evaluation (from Method 5)
   • Prototype learnings and validation results (from Method 6)
   • Constraint map and unresolved D/F/V gaps
   • Gap analysis with validated/assumed/unknown markers
   • Coaching context notes for the receiving RPI agent
5. Step 5: Generate RPI Entry — Create a structured handoff document at .copilot-tracking/dt/{project-slug}/rpi-handoff-solution-space.md that task-researcher or task-planner can consume directly as context.

Content Sanitization

The handoff artifact must:

• Remove internal coaching metadata not relevant to the RPI consumer
• Convert DT-specific terminology to RPI-understandable language where possible
• Preserve validated/assumed/unknown markers verbatim
• Include only actionable information — omit process notes and facilitator commentary

Token Budget

Target: ~800-1,200 tokens for the prompt template. Generated output size depends on tier.

How to Build This File

This is a .prompt.md file — use the prompt-builder agent (not task-implementor) for the authoring phase. The prompt-builder includes built-in Prompt Quality Criteria validation and sandbox testing specific to AI artifacts (.instructions.md, .prompt.md, .agent.md, SKILL.md).

Workflow: /task-research → /task-plan → /prompt-build → /task-review
Between each phase, run /clear to reset context.

Phase 1: Research

Gather source material for the Solution Space handoff prompt.

Source Material:

1. design-thinking-for-hve-capabilities/.github/chatmodes/design-thinking.chatmode.md (handoff patterns)
2. design-thinking-for-hve-capabilities/guidance/04-brainstorming.md
3. design-thinking-for-hve-capabilities/guidance/05-user-concepts.md
4. design-thinking-for-hve-capabilities/guidance/06-low-fidelity-prototypes.md

These files live in the DT4HVE repository. If you don't have local access, ask the user to provide them or use read_file if the repo is cloned nearby.

Steps:

1. Read all source materials above.
2. Read .github/instructions/prompt-builder.instructions.md for authoring standards.
3. Read any existing dt-handoff-* prompt files for structural precedent.
4. Gather content on Solution Space exit patterns, concept-to-implementation transition approaches, and analogous handoff examples from the cumulative research.

Starter prompt:

/task-research

Research for IS050: dt-handoff-solution-space.prompt.md

Read the DT4HVE source materials:
- design-thinking-for-hve-capabilities/.github/chatmodes/design-thinking.chatmode.md (handoff patterns and routing logic)
- design-thinking-for-hve-capabilities/guidance/04-brainstorming.md
- design-thinking-for-hve-capabilities/guidance/05-user-concepts.md
- design-thinking-for-hve-capabilities/guidance/06-low-fidelity-prototypes.md

Extract:
- Solution Space exit patterns — what signals readiness to leave Method 4/5/6
- Concept-to-implementation transition approaches — how DT artifacts map to RPI inputs
- Tiered exit schema for the Solution Space (which tier routes where)
- Analogous handoff examples from the cumulative research
- Artifact types carried forward: concept cards, D/F/V evaluations, prototype learnings, constraint maps

Also read .github/instructions/prompt-builder.instructions.md for authoring standards and any existing dt-handoff-*.prompt.md files for structural precedent.

Output: research summary from Phase 1 above

Phase 2: Plan

Plan the prompt structure and routing logic using the research output.

Steps:

1. Review the research output from Phase 1.
2. Plan the handoff prompt — context compilation strategy, target routing logic, tiered exit schema integration, state update behavior.
3. Define section ordering, token allocation, and confirm mode/tools frontmatter.

Starter prompt:

/task-plan

Plan for IS050: dt-handoff-solution-space.prompt.md

Using the Phase 1 research output, plan the handoff prompt:

- Handoff context compilation — which DT artifacts from Methods 4-6 feed into the prompt
- Target routing logic — how the prompt decides whether to route to task-researcher (feasibility gaps) or task-planner (ready for implementation)
- Tiered exit schema integration — maturity tiers and their routing destinations
- State update behavior — how the prompt updates DT session state on handoff
- Prompt mode: agent with tools read_file and create_file
- The prompt must read DT artifacts and produce a complete RPI-ready prompt file as output
- Token budget allocation (~800-1,200 tokens for the template)

Output: plan at .copilot-tracking/plans/{date}-is050-dt-handoff-solution-space-plan.md

Phase 3: Build

Author the prompt file using the prompt-builder agent.

Steps:

1. Review the plan from Phase 2.
2. Author the prompt file using /prompt-build.
3. The prompt must produce a complete RPI-ready prompt file that a developer can run immediately — not just a summary.

Starter prompt:

/prompt-build file=.github/prompts/dt-handoff-solution-space.prompt.md

Build IS050 using the plan at .copilot-tracking/plans/{date}-is050-dt-handoff-solution-space-plan.md.

This is a handoff prompt file for the Solution Space. Key authoring notes:
- Frontmatter: mode: agent, tools: ['read_file', 'create_file']
- The prompt reads DT artifacts (concept cards, D/F/V evaluations, prototype learnings, constraint maps) and produces a complete RPI-ready prompt file
- Routing logic based on tiered exit schema — routes to task-researcher for feasibility gaps or task-planner for implementation-ready concepts
- State update behavior — updates DT session state on handoff
- Carries forward all relevant Solution Space artifacts
- Output must be immediately usable as an RPI input — not a summary or checklist
- Token budget: ~800-1,200 tokens for the template

Phase 4: Review

Validate the prompt file against the plan and prompt-builder standards.

Steps:

1. Review the built file against prompt-builder standards and the issue requirements.
2. Validate handoff completeness, routing logic accuracy, tiered schema compliance, and prompt-builder standards.

Starter prompt:

/task-review

Review IS050: .github/prompts/dt-handoff-solution-space.prompt.md

Validate against:
- prompt-builder.instructions.md authoring standards
- Handoff completeness — all Solution Space artifacts carried forward
- Routing logic accuracy — correct tiered exit schema with appropriate destinations
- Tiered schema compliance — maturity tiers match the handoff contract
- State update behavior — session state is updated on handoff
- Frontmatter specifies mode: agent and tools: ['read_file', 'create_file']
- Output is a complete RPI-ready prompt file, not just a summary
- Token budget: ~800-1,200 tokens

Authoring Standards

Follow .github/instructions/prompt-builder.instructions.md:

• Mode: agent (needs read_file to compile DT artifacts)
• Prompt produces a .prompt.md file as output
• Handoff contract compliance is mandatory

Success Criteria

• File created at .github/prompts/dt-handoff-solution-space.prompt.md
• Frontmatter specifies mode: 'agent' and tools: ['read_file', 'create_file']
• Carries forward concept cards, D/F/V evaluations, prototype learnings, constraint maps
• Routes to task-researcher (feasibility gaps) or task-planner (ready for implementation) based on maturity
• Implements tiered exit schema from the handoff contract
• Applies validated-vs-unknown markers ([VALIDATED], [ASSUMED], [UNKNOWN])
• Updates coaching state on handoff execution
• Token count within ~800-1,200 for the prompt template
• 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