Overview

Create .github/prompts/dt-method-next.prompt.md, a routing prompt that assesses the current state of a Design Thinking project and recommends which method to work on next. Users invoke this prompt when they are unsure where they are in the methodology or need guidance on progression. The prompt reads project artifacts, evaluates completeness, and delegates to dt-coach for continued facilitation.

Target File

.github/prompts/dt-method-next.prompt.md

Frontmatter

---
description: 'Assess DT project state and recommend the next method to work on'
agent: 'dt-coach'
argument-hint: '[project-slug]'
---

Required Content

Prompt Purpose

When a user returns to a DT project after a break or is unsure of their progress, this prompt:

1. Locates the project artifacts in .copilot-tracking/dt/{project-slug}/
2. Inventories which method directories exist and their artifact completeness
3. Assesses the most recently active method
4. Recommends the next method based on methodology flow and artifact completeness

Input Variables

• ${input:project-slug}: (Required) The project identifier used in the artifact directory path. If omitted, the prompt lists available DT projects for user selection.

Method State Assessment

The prompt evaluates each method directory (01-09) for:

• Existence: Does the method directory exist?
• Artifact completeness: Does it contain the expected artifacts per the method's artifact specification?
• Checkpoint status: Are there checkpoint markers indicating method completion?

Assessment produces a status per method: Not Started, In Progress, or Complete.

Progression Logic

The methodology flows through three spaces:

The prompt recommends:

• The first incomplete method in sequence when previous methods are complete
• Returning to a previous method when artifacts suggest gaps
• Specific actions within the current method when it is in progress

Output Format

The prompt produces a brief project state summary and recommendation, then delegates to dt-coach for continued work:

## DT Project Status: {project-slug}

| Method       | Status      | Key Artifacts                                    |
| ------------ | ----------- | ------------------------------------------------ |
| 1 — Scoping  | Complete    | scoping-statement.md, stakeholder-map.md         |
| 2 — Research | In Progress | research-plan.md (interview guides missing)      |
| 3-9          | Not Started | —                                                |

**Recommendation**: Continue Method 2 — Research. Draft interview guides for the identified stakeholders.

After displaying the summary, the prompt activates dt-coach for the recommended method.

Edge Cases

• No project found: List available DT projects in .copilot-tracking/dt/ or suggest starting a new one with dt-coach
• All methods complete: Congratulate and suggest Method 9 handoff review or starting a new DT cycle
• Iteration loops: Methods 7-8 may loop; detect iteration cycle count and recommend progression to Method 9 when testing results stabilize

How to Build This File

This is a .prompt.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, use /clear to reset context, then attach the output artifact from the previous phase as input for the next.

Phase 1: Research

Research DT project artifact structure and method routing logic.

Source Material: DT coach and project structure: #file:.github/instructions/prompt-builder.instructions.md for authoring standards, #file:.github/agents/dt-coach.agent.md for the handoff delegation pattern, and DT project artifact directories for method state assessment.

Steps:

1. Type /clear to start a fresh conversation.
2. Attach #file:.github/instructions/prompt-builder.instructions.md and #file:.github/agents/dt-coach.agent.md.
3. Copy the prompt below into chat and send.

/task-research topic="DT method routing prompt"

Research the DT project artifact structure and method routing logic.

Extract:
- DT project artifact structure across all 9 methods
- Method sequencing rules, dependencies, and iteration patterns
- dt-coach.agent.md handoff delegation pattern
- Prompt-builder prompt file conventions (frontmatter, agent delegation, input variables)
- Method completeness indicators (what artifacts signal a method is done)
- Edge cases: no project, all complete, iteration loops in Methods 7-8

Output: DT method routing research

Phase 2: Plan

Plan the prompt: input variables, state assessment logic, and progression rules.

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 DT method routing prompt.

Use the attached research document as input. The plan should cover:
- Frontmatter structure (description, agent: 'dt-coach', argument-hint)
- Input variable (${input:project-slug})
- Method state assessment logic (read artifacts, evaluate completeness)
- Progression rules covering all 9 methods with correct sequencing
- Output format (project status summary + next method recommendation)
- Edge cases (no project, all complete, iteration loops)
- Delegation to dt-coach after recommendation

Output: .copilot-tracking/plans/{date}-dt-method-routing-plan.md

Phase 3: Build

Author the prompt file using prompt-builder.

Steps:

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

/prompt-build

Build the DT method routing prompt following the attached plan.

Create .github/prompts/dt-method-next.prompt.md with:
- Frontmatter: description, agent: 'dt-coach', argument-hint
- Input variable ${input:project-slug} documented
- Method state assessment reading artifact directories
- Progression logic covering all 9 methods with correct sequencing
- Output format with project status summary and recommendation
- Edge cases (no project found, all complete, iteration loops)
- Delegation to dt-coach after recommendation
- End with --- followed by activation instruction

The prompt must be a single-invocation workflow that assesses state and delegates.

Output: .github/prompts/dt-method-next.prompt.md

Phase 4: Review

Validate progression logic and prompt-builder compliance.

Steps:

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

/task-review

Review the method routing prompt against the attached plan.

Validate:
- Frontmatter includes description, agent: 'dt-coach', and argument-hint
- Input variable ${input:project-slug} documented
- Method state assessment reads artifact directories and evaluates completeness
- Progression logic covers all 9 methods with correct sequencing
- Output format provides clear project status summary and recommendation
- Edge cases handled (no project, all complete, iteration loops)
- Delegates to dt-coach after recommendation
- Prompt-builder compliance verified (--- activation, input variables)

Output: .copilot-tracking/reviews/{date}-dt-method-routing-review.md

After Review

• Pass — Open a PR with the prompt file.
• Iterate — Return to Phase 3 with the review document to fix identified issues.
• Escalate — Return to Phase 1 to investigate DT project structure gaps.

Authoring Standards

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

• Prompt file structure with agent frontmatter for delegation
• argument-hint shows expected input
• Input variables documented with ${input:} syntax
• End with --- followed by activation instruction

Success Criteria

• File created at .github/prompts/dt-method-next.prompt.md
• Frontmatter includes description, agent: 'dt-coach', and argument-hint
• Input variable ${input:project-slug} documented
• Method state assessment reads artifact directories and evaluates completeness
• Progression logic covers all 9 methods with correct sequencing
• Output format provides clear project status summary and recommendation
• Edge cases handled (no project, all complete, iteration loops)
• Delegates to dt-coach after recommendation
• Passes task-reviewer validation against prompt-builder standards
• Artifact registered in collections/design-thinking.collection.yml with path and kind fields
• Artifact registered in collections/hve-core-all.collection.yml with path and kind fields
• npm run plugin:generate run after updating collection manifests