Overview

Create dt-resume-coaching.prompt.md — the prompt users invoke to resume an existing DT coaching session. This prompt reads the coaching state file, provides a context summary of where the user left off, and re-establishes the coaching conversation at the correct method and phase.

Target File

.github/prompts/dt-resume-coaching.prompt.md

Frontmatter

---
description: 'Resume a Design Thinking coaching session — reads coaching state and re-establishes context'
mode: 'agent'
tools: ['read_file']
---

Required Steps

1. Step 1: Locate Project State — Find coaching state file at .copilot-tracking/dt/{project-slug}/coaching-state.md. If multiple projects exist, list them for user selection.
2. Step 2: Read and Summarize State — Parse coaching state: current method, phase, DT space, completed methods, hint calibration, and session recovery instructions.
3. Step 3: Context Recovery — Read recent method artifacts to build conversational context. Summarize: "Last session you were working on Method [N] ([name]), in the [phase] phase. Here's where you left off: [recovery instructions]."
4. Step 4: Resume Coaching — Hand off to dt-coach agent with restored context. Coach picks up the conversation naturally, not mechanically.

Input Variables

Prompt Behavior

The prompt:

• Reads coaching state from .copilot-tracking/dt/{project-slug}/coaching-state.md
• Reads the session_recovery_instructions field for free-text context
• Scans recent method artifact directories for last-modified content
• Provides human-readable summary of session position
• Restores hint calibration level so the coach doesn't reset escalation
• Hands off to dt-coach with full context — coach should feel like the same conversation continuing

Edge Cases

• No state file found: Suggest running dt-start-project prompt instead
• Multiple projects: List available projects with last-modified dates
• Corrupted state: Read what's available, flag missing fields, proceed with defaults

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

Phase 1: Research

Gather source material for the Resume Coaching prompt file.

Source Material: The upstream session recovery patterns — state parsing,
context rebuilding, hint calibration restoration, and coaching continuity — are
defined in the DT4HVE repository:

design-thinking-for-hve-capabilities/.github/chatmodes/design-thinking.chatmode.md

The chatmode defines how the coaching agent restores context from a previous
session, including state file parsing, progress summarization, and hint
calibration adjustment. Session recovery must handle edge cases like missing
state files, multiple projects, and corrupted state. If you don't have the
DT4HVE repository cloned, ask a maintainer for access.

Steps:

1. Type /clear to start a fresh conversation.
2. Attach the DT4HVE chatmode file listed above as context.
3. Copy the prompt below into chat and send.

/task-research topic="DT resume coaching prompt for hve-core"

Research the Design Thinking session recovery patterns for a .prompt.md file
targeting `.github/prompts/dt-resume-coaching.prompt.md`.

Use the attached DT4HVE Design Thinking Coach chatmode as the primary source.
Extract and analyze:
- Session recovery steps: state file location, parsing, validation
- Context rebuilding: how the coach summarizes prior progress for the user
- Hint calibration restoration: how coaching intensity adjusts based on
  session history
- Handoff to dt-coach agent after context is restored
- Edge cases: no state file found, multiple projects in `.copilot-tracking/dt/`,
  corrupted or partially written state

Focus areas:
- What information must be extracted from coaching-state.md to restore context
- How the context summary is presented to the user for confirmation
- How hint calibration carries forward between sessions
- How each edge case is detected and handled gracefully
- How the transition to dt-coach differs from a fresh start

The research output will feed `/task-plan` and then `/prompt-build` to author
the prompt file.

Output: DT resume coaching research

Phase 2: Plan

Plan the prompt file structure and recovery step sequencing using the research output.

Steps:

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

/task-plan

Plan the implementation of `.github/prompts/dt-resume-coaching.prompt.md` — a
prompt file for resuming Design Thinking coaching sessions.

Use the attached research document as the primary input.

The plan should cover:
- Frontmatter structure (`description`)
- Four steps: locate state, parse and validate, summarize context, hand off to
  dt-coach
- State parsing: extract current method, phase, progress, hint calibration
- Context summary: what to present to the user for confirmation
- Hint calibration restoration from session history
- Edge cases: no state file, multiple projects, corrupted state
- Transition to dt-coach after context is restored

The implementation phase will use `/prompt-build` (not task-implementor) since
this is an AI artifact file.

Output: .copilot-tracking/plans/{date}-dt-resume-coaching-plan.md

Phase 3: Build

Author the prompt file using the prompt-builder agent. Prompt-builder handles authoring standards (frontmatter, writing style, progressive disclosure) automatically through its built-in Prompt Quality Criteria validation.

Steps:

1. Type /clear to reset the conversation.
2. Attach the plan document from Phase 2 (find it in .copilot-tracking/plans/).
3. Copy the prompt below into chat and send.

/prompt-build file=.github/prompts/dt-resume-coaching.prompt.md

Build the DT Resume Coaching prompt file following the attached implementation
plan.

Requirements:
- Four steps: locate state, parse and validate, summarize context, hand off to
  dt-coach
- State parsing extracts current method, phase, progress, and hint calibration
- Context summary is presented to user for confirmation before proceeding
- Hint calibration restores coaching intensity from session history
- Edge cases handled: no state file (offer to start new project), multiple
  projects (prompt user to select), corrupted state (warn and offer recovery)
- Clean handoff to dt-coach agent with restored context
- Prompt follows prompt-builder standards for .prompt.md files

Output: .github/prompts/dt-resume-coaching.prompt.md

Phase 4: Review

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

Steps:

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

/task-review

Review the implementation of `.github/prompts/dt-resume-coaching.prompt.md`
against the attached plan.

Validate:
- Prompt Quality Criteria from `.github/instructions/prompt-builder.instructions.md`
- Four steps defined with clear sequencing
- State parsing correctly extracts all required fields from coaching-state.md
- Context summary is accurate and presented for user confirmation
- Hint calibration restoration preserves coaching intensity
- Edge cases covered: no state file, multiple projects, corrupted state
- Handoff to dt-coach is clean with restored context
- No temporal markers, task IDs, or DT4HVE-specific references remain
- Writing style follows prompt-builder standards for .prompt.md files

Output: .copilot-tracking/reviews/{date}-dt-resume-coaching-review.md

After Review

The reviewer reports one of three outcomes:

• Pass — The prompt file meets all criteria. Open a PR with the new file.
• Iterate — The reviewer identified specific issues. Return to Phase 3: /clear, attach the review document, then re-run /prompt-build with the fixes noted.
• Escalate — Fundamental gaps need more research. Return to Phase 1 to investigate the identified areas.

Success Criteria

• Prompt file created at .github/prompts/dt-resume-coaching.prompt.md
• Step-based protocol with 4 steps
• Reads coaching state and session recovery instructions
• Provides human-readable context summary
• Restores hint calibration level
• Handles edge cases (no state, multiple projects, corrupted state)
• Transition to dt-coach feels like continuing a conversation
• Passes task-reviewer validation against prompt-builder standards