# Pull Request

## Description

Creates `dt-method-02-deep.instructions.md`, an on-demand deep reference
for DT Method 2 (Design Research). The `dt-coach` agent loads this file
via `read_file` when users require advanced research guidance beyond
what the method-tier instruction provides.

The file contains 17 H3 subsections across 4 domains, each with prose,
bullets, and a labeled coaching prompt:

- **Advanced Interview Techniques** — pre-interview preparation,
cognitive interview techniques, silence and pause usage, reframing
resistance, cultural context adaptation
- **Ethnographic Observation Methods** — contextual inquiry setup,
shadow observation protocols, artifact analysis, environmental mapping
- **Evidence Triangulation** — multi-source validation, contradictory
evidence handling, pattern saturation assessment, insight confidence
scoring
- **Manufacturing Research Patterns** — shift-based research
considerations, production floor observation, operator vs. supervisor
perspectives, supply chain stakeholder mapping

Also adds `.github/agents/design-thinking` to `chat.agentFilesLocations`
in `.vscode/settings.json` so the `dt-coach` agent appears in the VS
Code agent dropdown.

## Related Issue(s)

Closes #585

## Type of Change

Select all that apply:

**Code & Documentation:**

- [ ] Bug fix (non-breaking change fixing an issue)
- [x] New feature (non-breaking change adding functionality)
- [ ] Breaking change (fix or feature causing existing functionality to
change)
- [ ] Documentation update

**Infrastructure & Configuration:**

- [ ] GitHub Actions workflow
- [ ] Linting configuration (markdown, PowerShell, etc.)
- [ ] Security configuration
- [ ] DevContainer configuration
- [ ] Dependency update

**AI Artifacts:**

- [x] Reviewed contribution with `prompt-builder` agent and addressed
all feedback
- [x] Copilot instructions (`.github/instructions/*.instructions.md`)
- [ ] Copilot prompt (`.github/prompts/*.prompt.md`)
- [ ] Copilot agent (`.github/agents/*.agent.md`)
- [ ] Copilot skill (`.github/skills/*/SKILL.md`)

**Other:**

- [ ] Script/automation (`.ps1`, `.sh`, `.py`)
- [ ] Other (please describe):

## Sample Prompts (for AI Artifact Contributions)

**User Request:**
> "We're struggling to get honest answers in interviews — people keep
telling us what they think we want to hear. How do we break through
that?"

**Execution Flow:**
1. `dt-coach` detects the user is in Method 2 (Design Research) with an
advanced interview challenge
2. Agent calls `read_file` on `dt-method-02-deep.instructions.md` to
load the deep reference
3. Agent surfaces the **Cognitive Interview Techniques** and **Reframing
Resistance** subsections
4. Agent responds using the coaching prompt: *"What would it look like
if participants felt safe enough to share what they actually think — and
what's currently in the way of that?"*

**Output Artifacts:**
No files created — the agent responds conversationally using the loaded
deep reference to coach the user through the research challenge.

**Success Indicators:**
- Agent surfaces subsection-specific coaching prompts rather than
generic research advice
- User receives technique-specific guidance (e.g., cognitive
interviewing, silence usage) matched to their specific challenge
- Coaching prompts feel contextual and non-prescriptive

## Testing

- Ran `/prompt-analyze` against `dt-method-02-deep.instructions.md` —
22/23 checks passed
- Reviewed and addressed all `prompt-builder` feedback via
`/prompt-build`
- Verified `applyTo` pattern matches method-tier convention
(`**/.copilot-tracking/dt/**/method-02*`)
- Confirmed collection manifests (`design-thinking`, `hve-core-all`)
updated and plugins regenerated
- Rebased cleanly onto updated `main` (post upstream sync of 106 files
including Methods 4–9)

## Checklist

### Required Checks

- [x] Documentation is updated (if applicable)
- [x] Files follow existing naming conventions
- [x] Changes are backwards compatible (if applicable)
- [ ] Tests added for new functionality (if applicable)

### AI Artifact Contributions

- [x] Used `/prompt-analyze` to review contribution
- [x] Addressed all feedback from `prompt-builder` review
- [x] Verified contribution follows common standards and type-specific
requirements

### Required Automated Checks

- [x] Markdown linting: `npm run lint:md` — ✅ 0 errors (181 files)
- [ ] Spell checking: `npm run spell-check` — ⚠️ pre-existing
environment issue (Node.js v20.16.0 < v20.18 required); unrelated to
this PR
- [x] Frontmatter validation: `npm run lint:frontmatter` — ✅ 0 errors, 0
warnings (167 files)
- [x] Skill structure validation: `npm run validate:skills` — ✅ 0 errors
(2 skills)
- [ ] Link validation: `npm run lint:md-links` — ⚠️ pre-existing
repo-wide ESM incompatibility in `markdown-link-extractor`; unrelated to
this PR
- [ ] PowerShell analysis: `npm run lint:ps` — ⚠️ 3 warnings in
`.copilot-tracking` sandbox script (`generate-dalle-images.ps1`), which
is gitignored and unrelated to this PR

## Security Considerations

- [x] This PR does not contain any sensitive or NDA information
- [x] Any new dependencies have been reviewed for security issues
- [x] Security-related scripts follow the principle of least privilege

## Additional Notes

The three automated check warnings are all pre-existing environment or
infrastructure issues:

1. **`spell-check`**: Requires Node.js ≥ 20.18; local environment is
20.16.0 — not introduced by this PR
2. **`lint:md-links`**: `markdown-link-extractor` ESM compatibility
error affects the entire repo — not introduced by this PR
3. **`lint:ps`**: Warnings are in
`.copilot-tracking/design-thinking-for-hve-capabilities/scripts/generate-dalle-images.ps1`,
a gitignored sandbox file — not introduced by this PR

Supersedes #677.

---------

Co-authored-by: Manasa Gudimella <mgudimella@microsoft.com>
Co-authored-by: Bill Berry <wbery@microsoft.com>