Overview

Create dt-method-05-concepts.instructions.md — the method-tier instruction file for Method 5: User Concepts. Users take the most promising ideas from brainstorming (Method 4) and develop them into articulated concepts with enough structure to evaluate. This method introduces the three-lens validation (Desirability / Feasibility / Viability) and M365 Copilot image prompt generation for concept visualization.

Target File

.github/instructions/dt-method-05-concepts.instructions.md

Frontmatter

---
description: 'Design Thinking Method 5: User Concepts — concept development, three-lens validation, and visual concept communication'
applyTo: '**/.copilot-tracking/dt/**/method-05*'
---

Required Content

Method Purpose

Method 5 transforms raw ideas into structured concepts. Users articulate what the idea is, who it serves, and why it matters. Each concept is evaluated through three lenses — Desirability (do users want this?), Feasibility (can we build this?), Viability (should we build this?) — to identify concepts worth prototyping.

Three Sub-Methods

Two Specialized Hats

Three-Lens Validation (D/F/V)

Each concept is evaluated through:

• Desirability: Does this address a real user need? Is there evidence from Method 2 research? Would users choose this over alternatives?
• Feasibility: Can this be built with available technology, skills, and time? What are the technical risks? What unknowns remain?
• Viability: Does this align with organizational goals? Is there a sustainable model? What's the cost-to-value ratio?

The coach helps users hold all three lenses simultaneously without letting any single lens dominate. A concept strong in desirability but weak in feasibility isn't discarded — it's flagged for feasibility research in later methods.

M365 Copilot Image Prompt Generation

The coach produces descriptive image prompts that users copy-paste into M365 Copilot for concept visualization:

• Prompts describe the concept scenario visually — who is using it, what they see, what the environment looks like
• Image prompts are framed for communication, not for wireframing (lo-fi constraint still applies)
• Example structure: "Create an image showing [persona] in [context] using [concept] to [accomplish goal]. The style should be [sketch/illustration] to convey that this is an early concept, not a final design."
• The coach does not generate images directly — it produces prompts the user takes to M365 Copilot

Method 5 Artifacts

Outputs stored at .copilot-tracking/dt/{project-slug}/method-05-concepts/:

• concept-cards/ — One file per concept: description, value proposition, target user, D/F/V assessment
• dfv-matrix.md — Comparative D/F/V evaluation across concepts
• image-prompts.md — M365 Copilot image prompts for concept visualization
• selection-summary.md — Concepts selected for prototyping with rationale

Lo-Fi Quality Enforcement

Method 5 increases fidelity slightly from Method 4 but remains conceptual:

• Concept cards are structured narratives, not specification documents
• D/F/V assessments are conversational evaluations, not scored rubrics
• Image prompts describe scenarios, not interface layouts
• Coach redirects over-specification: "That's getting into implementation — let's keep it at the concept level for now"

Coaching Approach

The coach in Method 5:

• Bridges ideation to structure: "You have a great raw idea — let's figure out what makes it tick"
• Challenges vague concepts: "Who specifically would use this, and what would change for them?"
• Balances optimism with realism through the three lenses without being a gatekeeper
• Encourages visual communication: "Want an image prompt you can use in Copilot to show what this concept looks like in practice?"

Token Budget

Target: ~1,500-2,000 tokens (method tier)

How to Build This File

This is an .instructions.md file — use the prompt-builder agent (not task-implementor) for the authoring phase.

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

Phase 1: Research

Source Material: design-thinking-for-hve-capabilities/guidance/05-user-concepts.md
This file lives in the DT4HVE repository. If you don't have local access, ask the user to provide it or use read_file if the repo is cloned nearby.

Steps:

1. Read the source material above.
2. Read .github/instructions/prompt-builder.instructions.md for authoring standards.
3. Read any existing dt-method-* instruction files for structural precedent.
4. Gather content on concept articulation patterns, three-lens D/F/V evaluation framework, concept card generation, and M365 Copilot image prompt integration.

Starter prompt:

/task-research

Research for IS028: dt-method-05-concepts.instructions.md

Read the DT4HVE source material at design-thinking-for-hve-capabilities/guidance/05-user-concepts.md. Extract:
- Concept articulation patterns — how raw ideas become structured concepts
- Three-lens evaluation framework (Desirability / Feasibility / Viability)
- Concept card generation — structure, content, and quality expectations
- M365 Copilot image prompt integration — how the coach produces visualization prompts
- Lo-fi quality enforcement for concept-level artifacts

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

Output: research summary from Phase 1 above

Phase 2: Plan

Steps:

1. Review the research output from Phase 1.
2. Plan the instruction file structure — method purpose, three sub-methods, two specialized hats, D/F/V framework, concept cards, image prompt guidance.
3. Define section ordering, token allocation, and applyTo targeting.

Starter prompt:

/task-plan

Plan for IS028: dt-method-05-concepts.instructions.md

Using the Phase 1 research output, plan the instruction file:

- Method purpose: transforming raw ideas into structured, evaluable concepts
- Three sub-methods: 5a Concept Planning, 5b Concept Articulation, 5c Concept Evaluation
- Two hats: Concept Architect, Three-Lens Evaluator
- D/F/V framework: balanced evaluation that is facilitative, not judgmental
- Concept cards: structured narratives with value propositions and target users
- Image prompt guidance: M365 Copilot prompts as a natural coaching behavior
- Section ordering and token budget allocation (~1,500-2,000 tokens)
- applyTo: '**/.copilot-tracking/dt/**/method-05*'

Output: plan at .copilot-tracking/plans/{date}-is028-dt-method-05-plan.md

Phase 3: Build

Steps:

1. Review the plan from Phase 2.
2. Author the instruction file using /prompt-build.
3. Integrate image prompt generation as a natural coaching behavior. Ensure D/F/V evaluation is facilitative, not judgmental.

Starter prompt:

/prompt-build file=.github/instructions/dt-method-05-concepts.instructions.md

Build IS028 using the plan at .copilot-tracking/plans/{date}-is028-dt-method-05-plan.md.

This is a method-tier instruction file for Method 5: User Concepts. Key authoring notes:
- applyTo targets Method 5 artifact paths: '**/.copilot-tracking/dt/**/method-05*'
- Integrate image prompt generation as a natural coaching behavior, not a separate tool
- Ensure D/F/V evaluation is facilitative, not judgmental — the coach helps users hold all three lenses
- Three sub-methods with consistent table structure
- Two specialized hats with activation triggers
- Concept cards are structured narratives, not specification documents
- Lo-fi enforcement: concepts are articulated but not over-specified
- M365 Copilot image prompts are guidance for the user, not AI self-generation

Phase 4: Review

Steps:

1. Review the built file against prompt-builder standards and the issue requirements.
2. Validate D/F/V framework completeness, concept card quality, image prompt integration, and coaching facilitation tone.

Starter prompt:

/task-review

Review IS028: .github/instructions/dt-method-05-concepts.instructions.md

Validate against:
- prompt-builder.instructions.md authoring standards
- D/F/V framework completeness — all three lenses with balanced evaluation guidance
- Concept card quality — structured narratives with clear value propositions
- Image prompt integration — M365 Copilot prompts woven into coaching, not bolted on
- Coaching facilitation tone — evaluative but not judgmental
- Lo-fi enforcement for concept artifacts
- Token budget: ~1,500-2,000 tokens
- Frontmatter applyTo correctness
- Three sub-methods and two hats with consistent table structure

After Review

• Pass: Mark IS028 complete.
• Iterate: Address review findings, rebuild, re-review.
• Escalate: If blocked by missing DT4HVE source material or architectural questions, raise to the user.

Authoring Standards

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

• applyTo targets Method 5 artifact paths
• Three sub-methods with consistent table structure
• Two specialized hats with activation triggers
• M365 Copilot image prompts are guidance for the user, not AI self-generation

Success Criteria

• File created at .github/instructions/dt-method-05-concepts.instructions.md
• Frontmatter applyTo targets Method 5 artifact paths
• Three sub-methods defined (planning, articulation, evaluation)
• Two specialized hats with clear activation triggers
• Three-lens validation (D/F/V) framework with coaching guidance for balanced evaluation
• M365 Copilot image prompt generation integrated as a coaching behavior
• Lo-fi quality constraints enforced — concepts are structured narratives, not specs
• Token count within ~1,500-2,000 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