Dt Rpi Integration — Agent Skill by Microsoft

From the author

Design Thinking handoff knowledge for research-ready rpi-research inputs and DT-aware rpi-plan, rpi-implement, and rpi-review context

— the description by Microsoft

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

This skill acts as the bridge between Design Thinking coaching and the RPI (Research, Plan, Implement, Review) workflow. It defines the formal handoff contracts, artifact schemas, and contextual framing required to ensure that insights gathered during Design Thinking are effectively utilized by downstream RPI phases. It ensures that regardless of the Design Thinking exit point, all projects enter the RPI pipeline through the research phase with standardized, validated documentation.

What it does

Defines the formal handoff contract for transitioning from Design Thinking to RPI phases.
Provides standardized schemas for documenting handoff artifacts and project constraints.
Offers guidance on framing research, planning, implementation, and review contexts based on DT artifacts.
Includes procedures for subagent dispatch to validate handoff readiness.
Provides specific instructions for generating lo-fi image prompts for concept visualization.

When to use it

When a team graduates from a Design Thinking space boundary and needs to begin the RPI workflow.
When you need to prepare research-ready inputs from completed Design Thinking methods.
When you need to ensure that RPI planning and implementation phases are informed by validated DT context.
When you need to generate neutral, lo-fi concept sketches for stakeholder evaluation.

Advantages

Ensures continuity of project context across different development phases.
Standardizes the quality and format of handoff documentation.
Reduces investigation scope by providing validated artifacts from earlier stages.
Maintains neutrality in visual concept generation through strict prompt enforcement.

Usage examples

“How do I structure the handoff summary when moving from the Problem space to RPI research?”: The agent provides the required YAML schema for the handoff-summary.md file, including fields for exit points, artifacts, and constraint tracking.
“Generate a lo-fi image prompt for a concept where a technician uses a tablet to check inventory.”: The agent generates a prompt following the required stick-figure template, ensuring no personal traits are included and the style is set to black-and-white line art.

Install

Download the skill as a ZIP:
https://mp-staging.autogineer.com/agent-skills/microsoft/dt-rpi-integration/download

Unzip it into the cross-client skills folder of your project:

mkdir -p .agents/skills && unzip dt-rpi-integration.zip -d .agents/skills

Documentation by Microsoft

Design Thinking → RPI Integration — Skill Entry

This skill is the entry point for Design Thinking to RPI integration knowledge.

The DT coach loads these references when Design Thinking coaching graduates into the RPI workflow. Every DT exit produces research-ready input for rpi-research. The retained downstream phases, rpi-plan, rpi-implement, and rpi-review, consume the resulting context. RPI Agent is the lifecycle wrapper when end-to-end coordination is needed; it does not replace rpi-research as the handoff target.

Integration references

Reference	When to load
Handoff contract	Exit points, artifact schemas, RPI input contracts, and quality markers for lateral DT-to-RPI handoff
Space-exit handoffs	Step-by-step Problem, Solution, and Implementation Space exit procedures that compile the handoff summary and the RPI entry document
Research context	DT-aware `rpi-research` framing for handoffs from the DT coach
Planning context	DT-aware `rpi-plan` context for plans originating from DT artifacts
Implement context	DT-aware `rpi-implement` context applying fidelity and stakeholder constraints
Review context	DT-aware `rpi-review` criteria for evaluating Design Thinking artifacts
Subagent handoff	Readiness assessment, artifact compilation, and validation via subagent dispatch
Image prompt generation	Method 5 concept visualization with lo-fi prompt enforcement

Skill layout

SKILL.md — this file (skill entrypoint).
references/ — the DT-to-RPI integration reference documents.
- rpi-handoff-contract.md — DT-to-RPI handoff contract: exit points, artifact schemas, RPI input contracts, and confidence markers.
- space-exit-handoffs.md provides the per-space exit procedures, handoff summaries, transition entries, and RPI entry documents.
- rpi-research-context.md provides DT-aware rpi-research context.
- rpi-planning-context.md provides DT-aware rpi-plan context.
- rpi-implement-context.md provides DT-aware rpi-implement context.
- rpi-review-context.md provides DT-aware rpi-review context.
- subagent-handoff.md provides the subagent dispatch workflow for handoff readiness, compilation, and validation.
- image-prompt-generation.md provides Method 5 lo-fi image prompt generation guidance.

Files

ATTRIBUTION.mdAdded by AutoGineer887 B
LICENSELicence file added by AutoGineer1.1 KiB
SKILL.md3.7 KiB
references/image-prompt-generation.md9.0 KiB
references/rpi-handoff-contract.md7.2 KiB
references/rpi-implement-context.md4.3 KiB
references/rpi-planning-context.md5.8 KiB
references/rpi-research-context.md4.2 KiB
references/rpi-review-context.md6.1 KiB
references/space-exit-handoffs.md17.3 KiB
references/subagent-handoff.md3.3 KiB

references/rpi-review-context.md

title: 'DT Review Context' description: Criteria augmenting rpi-review when evaluating Design Thinking artifacts for coaching quality and method fidelity.

When rpi-review evaluates DT-origin work, these criteria augment the standard review workflow. The review phase receives DT context through rpi-research, rpi-plan, and rpi-implement, not through a direct DT exit. The review question shifts from "does the code work?" to "does the artifact serve the Design Thinking process?" Evaluate coaching quality, method fidelity, and stakeholder coverage alongside structural correctness.

Review Criteria Adjustments

Standard Review	DT-Informed Review
Code correctness focus	Coaching quality and method fidelity focus
Pass/fail assessment	Space-appropriate fidelity assessment (rough/scrappy/functional)
Style guide conformance	Coaching identity conformance (Think/Speak/Empower)
Single output evaluation	Multi-stakeholder coverage evaluation
Forward-only approval	Iteration-aware evaluation supporting return paths

Quality Criteria by Artifact Type

Artifact Type	Key Quality Criteria
Coaching instructions	Think/Speak/Empower structure; coaching boundaries maintained; no directive language; progressive hints
Method instructions	Correct space assignment; exit signals defined; coaching hat triggers; non-linear iteration support
Deep instructions	Advanced techniques beyond base method; domain expertise depth; fidelity appropriate to space
Industry context	Domain vocabulary mapping; industry-specific constraints; stakeholder archetypes; reference scenarios
Handoff artifacts	Confidence markers applied (validated/assumed/unknown/conflicting); exit point and target phase alignment
Agent definitions	Subagent delegation patterns; handoff labels; core principles aligned with coaching identity
Method output artifacts	Fidelity matches current space; multi-source evidence; stakeholder coverage across identified groups
Coaching state artifacts	Session continuity maintained; method progress accurately tracked; recovery points clearly marked; no stale references

DT Review Checklist Additions

When reviewing DT artifacts, add these checks to the standard review checklist:

Language guides thinking through observations ("I'm noticing..."), not directives ("You must..."), to maintain coaching tone
Output quality matches the method's space tier: rough in Problem, scrappy in Solution, functional in Implementation
Perspectives from all identified stakeholder groups are represented, not just the most obvious
Claims trace to research data (Method 2+) or acknowledged assumptions (Method 1) for evidence grounding
Content encourages revisiting and refining without framing backward movement as regression
Token budgets follow artifact limits: ambient (coaching identity, quality constraints, review context) ≤800; method ≤1,500; deep ≤2,500 tokens

Anti-Patterns to Flag

Anti-Pattern	Severity	Rationale
Directive coaching language ("You must...")	Major	Violates Think/Speak/Empower identity
Production-quality output in early methods	Major	Violates anti-polish stance and space fidelity rules
Missing stakeholder perspectives	Major	Violates multi-stakeholder requirement
Single-source conclusions	Major	Violates multi-source validation rule
Skipped method exit signals	Critical	Invalidates downstream work; violates method sequencing
Confidence markers missing from handoff artifacts	Major	Downstream RPI phases cannot assess artifact reliability
Unresolved conflicting markers passed downstream	Critical	Invalidates downstream work; violates handoff contract reliability

Severity Mapping

Severity	Description	Examples
Critical	Violations that invalidate downstream work	Skipped exit signals, wrong space assignment, unresolved conflicting markers passed downstream
Major	Violations that degrade artifact quality	Directive language, missing stakeholders, single-source conclusions, missing confidence markers, over-polished prototypes
Minor	Stylistic or structural issues	Leaked internal reasoning in Speak layer, ideal-only testing conditions, token budget slightly exceeded

These criteria complement co-loaded instruction files (dt-quality-constraints, dt-coaching-identity, dt-method-sequencing, dt-rpi-handoff-contract): reference their content during review rather than duplicating it.

All DT coaching artifacts are scoped to .copilot-tracking/dt/{project-slug}/. Never write DT artifacts directly under .copilot-tracking/dt/ without a project-slug directory.

Requirements and compatibility

Written for: GitHub Copilot, Claude Code

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Needs: Claude Code frontmatter

Security

No issues found by our static scan· No issues found

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Handoff
Matt Pocock· Community
Generates a concise handoff document to facilitate seamless transitions between AI agents.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository317.5k
Chief Of Staff
Matt Pocock· Community
A strategic orchestration agent that manages long-term goals by coordinating subagents and maintaining a persistent knowledge wiki.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository317.5k
Ask Matt
Matt Pocock· Community
A central routing skill that helps you identify the correct agent skill or workflow for your current situation.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository317.5k
Claude Handoff
Matt Pocock· Community
Seamlessly transfer your current conversation context to a new background agent.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository317.5k
Executing Plans
Jesse Vincent (obra)· Community
Execute implementation plans inline without subagents, maintaining rigorous progress tracking and verification.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
MITLow riskHas scriptsGitHub stars of the source repository314k
Using Superpowers
Jesse Vincent (obra)· Community
A mandatory protocol for AI agents to identify and invoke relevant skills before performing any task or responding to user input.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository314k

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download dt-rpi-integration.zip