Overview

Create collections/design-thinking.collection.yml — the collection manifest for all Design Thinking artifacts. The collection system (PR #439) uses this manifest to discover, validate, and package DT artifacts for distribution. This file is the single source of truth for what constitutes the "Design Thinking" collection.

This is the starter issue for Phase 6 — no intra-phase dependencies, but all Phases 1-5 artifacts must exist before the manifest can list them.

Target File

collections/design-thinking.collection.yml

Required Content

Collection Metadata

Top-level metadata fields:

• id: design-thinking
• name: Design Thinking
• description: Design Thinking methodology coach and learning tools for HVE Core — 9 methods from research through implementation with AI-assisted facilitation
• tags: design-thinking, methodology, coaching, facilitation, innovation

Artifact Inventory

The manifest lists every DT artifact by category and relative path:

Each entry includes:

• path: Relative path from repository root
• kind: One of agent, instruction, prompt, skill

Manifest Structure Pattern

Follow the pattern established by existing collection manifests in the collections/ directory. Reference ado.collection.yml or rpi.collection.yml for structural conventions — item ordering, field names, and YAML formatting.

Validation Integration

The manifest must pass the existing collection validation pipeline:

• npm run plugin:generate regenerates plugin outputs from the manifest
• npm run lint:yaml validates YAML syntax
• npm run lint:frontmatter validates any generated markdown frontmatter

How to Build This File

This is a .collection.yml manifest — use the task-implementor agent (not prompt-builder) for the authoring phase.

Workflow: /task-research → /task-plan → /task-implement → /task-review

Between each phase, use /clear to reset context, then attach the output artifact from the previous phase as input for the next.

Phase 1: Research

Gather structural patterns from existing collection manifests.

Source Material: Existing collection manifests in collections/ — reference rpi.collection.yml and ado.collection.yml for field conventions, item schema, and YAML formatting.

Steps:

1. Type /clear to start a fresh conversation.
2. Attach #file:collections/rpi.collection.yml and #file:collections/ado.collection.yml as context.
3. Copy the prompt below into chat and send.

/task-research topic="Design Thinking collection manifest"

Research the collection manifest format for a new design-thinking.collection.yml.

Using the attached collection manifests as references, extract:
- Top-level metadata fields (id, name, description, tags)
- Item entry schema (path, kind) and valid kind values
- Item ordering conventions and YAML formatting patterns
- Display section options (ordering, featured)
- How maturity flags are used for experimental items

Then inventory all Design Thinking artifacts by scanning:
- .github/agents/dt-*.agent.md
- .github/instructions/dt-*.instructions.md
- .github/prompts/dt-*.prompt.md

Output: DT collection manifest research

Phase 2: Plan

Plan the manifest structure and complete artifact inventory.

Steps:

1. Type /clear to reset the conversation.
2. Attach the research document from Phase 1.
3. Copy the prompt below into chat and send.

/task-plan

Plan the implementation of collections/design-thinking.collection.yml.

Use the attached research document as input. The plan should cover:
- Collection metadata (id, name, description, tags)
- Complete item listing with path and kind for every DT artifact
- Item ordering strategy (by category: agents, then prompts, then instructions)
- Display section configuration
- Validation commands to run after creation

Output: .copilot-tracking/plans/{date}-dt-collection-manifest-plan.md

Phase 3: Implement

Create the YAML manifest following the established pattern.

Steps:

1. Type /clear to reset the conversation.
2. Attach the plan document from Phase 2 and #file:collections/rpi.collection.yml as a formatting reference.
3. Copy the prompt below into chat and send.

/task-implement

Implement collections/design-thinking.collection.yml following the attached
plan.

Create the YAML manifest listing every Design Thinking artifact with correct
paths and kind values. Match the formatting conventions from peer manifests
(indentation, comment grouping, field ordering).

After creating the file, run:
- npm run lint:yaml to validate YAML syntax
- npm run plugin:generate to validate collection pipeline integration

Output: collections/design-thinking.collection.yml

Phase 4: Review

Validate manifest completeness and pipeline compliance.

Steps:

1. Type /clear to reset the conversation.
2. Attach the plan document from Phase 2 and the new manifest file from Phase 3.
3. Copy the prompt below into chat and send.

/task-review

Review collections/design-thinking.collection.yml against the attached plan.

Validate:
- Every DT artifact from Phases 1-5 and 7 is listed with correct path and kind
- Top-level metadata matches conventions from peer manifests
- YAML formatting follows established patterns
- npm run plugin:generate succeeds
- npm run lint:yaml passes
- No artifacts are missing from the inventory

Output: .copilot-tracking/reviews/{date}-dt-collection-manifest-review.md

After Review

• Pass — Open a PR with the new manifest.
• Iterate — Return to Phase 3 with the review document to fix identified issues.
• Escalate — Return to Phase 1 to investigate gaps in artifact discovery.

Authoring Standards

Follow existing collection manifest conventions:

• Match field names (path, kind) and ordering from peer manifests
• Use consistent YAML formatting (indentation, quoting)
• Include all DT artifacts — omitting an artifact excludes it from distribution

Success Criteria

• File created at collections/design-thinking.collection.yml
• Collection metadata fields populated (id, name, description, tags)
• Every DT agent, instruction, and prompt artifact listed with correct paths and kind values
• npm run plugin:generate succeeds with the new manifest
• npm run lint:yaml passes
• Manifest structure matches conventions from peer collection files