Dt Methods
MicrosoftOfficialPublished by Microsoft
- Skill
- Guidelines
- Business and productivity
- MIT
- No issues found
AutoGineer summary
Written by AutoGineer from the skill’s own files, not by the author
The dt-methods skill provides a structured reference system for AI agents to coach teams through the Design Thinking process. It organizes guidance into three layers: core method references for day-to-day activities, deep references for advanced facilitation and recovery, and industry-specific context to tailor vocabulary and constraints. The skill covers nine major industries, including energy, healthcare, and financial services, ensuring that coaching is grounded in the specific operational realities and regulatory environments of the user's sector.
What it does
- Provides step-by-step guidance for nine core Design Thinking methods.
- Offers advanced facilitation techniques and recovery strategies for complex scenarios.
- Maps Design Thinking concepts to industry-specific terminology and constraints.
- Integrates telemetry expectations for session artifacts to ensure observable outcomes.
- Supplies industry-specific empathy tools and stakeholder considerations.
- Supports tailoring of coaching based on the active method and identified industry context.
When to use it
- When facilitating a Design Thinking session and requiring method-specific guidance.
- When a team needs advanced troubleshooting or recovery techniques during a workshop.
- When applying Design Thinking within a specific industry to ensure relevant vocabulary and constraints are respected.
- When defining telemetry or audit trails for Design Thinking session artifacts.
Advantages
- Ensures consistent coaching across the entire Design Thinking lifecycle.
- Reduces friction by using industry-appropriate language and concepts.
- Provides a clear path for scaling from basic methods to advanced facilitation.
- Integrates operational and regulatory awareness into the design process.
Usage examples
- “How should I facilitate a brainstorming session for a team in the energy sector?”
- The agent will load the brainstorming method reference and the energy industry context, providing guidance that uses energy-specific terms like 'reliability concern' and 'forced outage rate' while emphasizing safety and regulatory constraints.
- “We are stuck during the user research phase; what advanced techniques can we use?”
- The agent will access the deep reference for the Design Research method to provide advanced facilitation strategies and recovery techniques to help the team overcome their current roadblock.
Install
Download the skill as a ZIP:
https://mp-staging.autogineer.com/agent-skills/microsoft/dt-methods/downloadUnzip it into the cross-client skills folder of your project:
mkdir -p .agents/skills && unzip dt-methods.zip -d .agents/skills
Documentation by Microsoft
Design Thinking Methods Skill
Entrypoint for Design Thinking method coaching knowledge. The dt-coach agent loads the method reference matching the active method in coaching state, the matching deep reference when advanced technique is needed, and the industry reference when an industry context applies.
How to use this skill
Read the method reference for the method currently active in coaching state to ground day-to-day coaching.
Read the matching deep reference when the conversation needs advanced facilitation, recovery techniques, or expert frameworks beyond the core method guidance.
Read the industry reference when the team has identified one of the nine covered industries as their context, and weave its vocabulary, constraints, and empathy tools into method-specific guidance.
Read dt-coach-telemetry.md when DT session artifacts need observable telemetry expectations grounded in
telemetry-foundations.
Method references
| Method | Name | Core reference | Deep reference |
|---|---|---|---|
| 1 | Scope Conversations | method-01-scope.md | method-01-deep.md |
| 2 | Design Research | method-02-research.md | method-02-deep.md |
| 3 | Input Synthesis | method-03-synthesis.md | method-03-deep.md |
| 4 | Brainstorming | method-04-brainstorming.md | method-04-deep.md |
| 5 | User Concepts | method-05-concepts.md | method-05-deep.md |
| 6 | Low-Fidelity Prototypes | method-06-lofi-prototypes.md | method-06-deep.md |
| 7 | High-Fidelity Prototypes | method-07-hifi-prototypes.md | method-07-deep.md |
| 8 | User Testing | method-08-testing.md | method-08-deep.md |
| 9 | Iteration at Scale | method-09-iteration.md | method-09-deep.md |
Industry context references
Load the matching reference when the team identifies its industry context.
| Reference | When to load |
|---|---|
| industry-energy.md | Energy-sector vocabulary, constraints, empathy tools, and reference scenario. |
| industry-financial-services.md | Financial services vocabulary, constraints, empathy tools, and reference scenario. |
| industry-healthcare.md | Healthcare vocabulary, constraints, empathy tools, and reference scenario. |
| industry-manufacturing.md | Manufacturing vocabulary, constraints, empathy tools, and reference scenario. |
| industry-nonprofit-social-impact.md | Nonprofit and social impact vocabulary, constraints, empathy tools, and reference scenario. |
| industry-pharmaceuticals-life-sciences.md | Pharmaceuticals and life sciences vocabulary, constraints, empathy tools, and reference scenario. |
| industry-professional-services.md | Professional services vocabulary, constraints, empathy tools, and reference scenario. |
| industry-public-sector.md | Public sector and government vocabulary, constraints, empathy tools, and reference scenario. |
| industry-retail-cpg.md | Retail and consumer goods vocabulary, constraints, empathy tools, and reference scenario. |
Skill layout
.
├── SKILL.md
└── references/
├── method-01-scope.md
├── method-01-deep.md
├── method-02-research.md
├── method-02-deep.md
├── method-03-synthesis.md
├── method-03-deep.md
├── method-04-brainstorming.md
├── method-04-deep.md
├── method-05-concepts.md
├── method-05-deep.md
├── method-06-lofi-prototypes.md
├── method-06-deep.md
├── method-07-hifi-prototypes.md
├── method-07-deep.md
├── method-08-testing.md
├── method-08-deep.md
├── method-09-iteration.md
├── method-09-deep.md
├── industry-energy.md
├── industry-financial-services.md
├── industry-healthcare.md
├── industry-manufacturing.md
├── industry-nonprofit-social-impact.md
├── industry-pharmaceuticals-life-sciences.md
├── industry-professional-services.md
├── industry-public-sector.md
├── industry-retail-cpg.md
└── dt-coach-telemetry.mdFiles
- ATTRIBUTION.mdAdded by AutoGineer871 B
- LICENSELicence file added by AutoGineer1.1 KiB
- SKILL.md6.5 KiB
- references/dt-coach-telemetry.md1.5 KiB
- references/industry-energy.md9.1 KiB
- references/industry-financial-services.md10.7 KiB
- references/industry-healthcare.md7.7 KiB
- references/industry-manufacturing.md6.7 KiB
- references/industry-nonprofit-social-impact.md15.6 KiB
- references/industry-pharmaceuticals-life-sciences.md20.1 KiB
- references/industry-professional-services.md14.0 KiB
- references/industry-public-sector.md12.1 KiB
- references/industry-retail-cpg.md16.5 KiB
- references/method-01-deep.md10.4 KiB
- references/method-01-scope.md10.2 KiB
- references/method-02-deep.md15.6 KiB
- references/method-02-research.md7.0 KiB
- references/method-03-deep.md13.9 KiB
- references/method-03-synthesis.md8.9 KiB
- references/method-04-brainstorming.md13.6 KiB
- references/method-04-deep.md16.5 KiB
- references/method-05-concepts.md17.1 KiB
- references/method-05-deep.md17.7 KiB
- references/method-06-deep.md16.3 KiB
- references/method-06-lofi-prototypes.md17.2 KiB
- references/method-07-deep.md10.9 KiB
- references/method-07-hifi-prototypes.md9.8 KiB
- references/method-08-deep.md15.0 KiB
- references/method-08-testing.md11.5 KiB
- references/method-09-deep.md17.2 KiB
- references/method-09-iteration.md11.9 KiB
references/method-03-synthesis.md
title: 'DT Method 03: Input Synthesis' description: Design Thinking Method 03 (Input Synthesis) for clustering research signals into insights and opportunity areas.
Input synthesis aggregates research inputs (interviews, surveys, reports, observations) to find patterns, themes, and insights that form a complete picture of the problem. Without proper synthesis, teams move to brainstorming with fragmented understanding instead of unified problem clarity.
Purpose
Transform fragmented research data from diverse sources into unified problem understanding that sets clear direction for solution development.
Sub-Method Phases
Phase 1: Synthesis Planning
Organize all Method 2 research outputs and establish the synthesis strategy. Determine data source availability, identify coverage gaps, and define the analysis approach.
Exit criteria: an input inventory exists cataloging all Method 2 artifacts with source type, coverage assessment, and a defined synthesis approach.
Phase 2: Synthesis Execution
Perform systematic pattern recognition across organized inputs. Identify cross-source themes, validate patterns through multiple perspectives, and develop unified insights.
Exit criteria: affinity clusters and insight statements exist with each theme supported by evidence from multiple independent sources.
Phase 3: Synthesis Documentation
Formalize validated patterns into structured problem definitions, insight statements, and how-might-we questions. Run synthesis validation across all five dimensions before declaring transition readiness.
Exit criteria: problem definition and how-might-we artifacts exist, five-dimension validation shows strength across all dimensions, and the team confirms shared problem understanding.
Coaching Hats
Two specialized hats activate based on conversation context. See method-03-deep.md for detailed hat guidance, activation triggers, coaching focus areas, and Human-AI Collaboration patterns.
| Hat | Role | Activation |
|---|---|---|
| Pattern Analyst | Data organization, cross-source theme discovery, evidence-based validation | Raw data shared, clustering needed, contradictions detected |
| Insight Framer | Problem articulation, HMW questions, transition readiness assessment | Patterns validated, problem framing needed, transition discussion |
Pattern Recognition Framework
Data Source Integration
Combine input sources: worker interviews revealing capability constraints, observational data showing environmental factors, performance reports quantifying impact metrics
Look for themes that emerge when different data sources confirm and reinforce each other
Theme Development
Individual insight: a specific observation or quote from one source
Supporting evidence: confirmation from additional sources and perspectives
Unified theme: the pattern connecting all perspectives into a coherent problem statement
Actionable direction: framing that guides solution development without prescribing specific solutions
Systematic Synthesis Process
Input Organization
Stakeholder inputs: interview transcripts, conversation notes, survey responses, stakeholder mapping
Observational data: workflow documentation, process mapping, environmental constraint analysis
Quantitative data: performance metrics, operational analytics, resource utilization
Pattern Recognition
Validate patterns through different stakeholder perspectives
Connect qualitative observations with quantitative metrics
Require multiple independent sources to support each theme before considering it robust
Unified Insight Development
Prioritize themes based on stakeholder impact and solution potential
Frame insights as actionable direction without prescribing specific solutions
Validate synthesis against original research data for accuracy
Synthesis Validation
Before transitioning to brainstorming, evaluate synthesis quality across five dimensions.
Research fidelity: synthesis accurately reflects collected evidence rather than assumptions
Stakeholder completeness: themes include the full range of relevant stakeholder groups
Pattern robustness: patterns appear across multiple data points, not from isolated anecdotes
Actionability: outputs translate into clear problem statements that can guide solution work
Team alignment: the team shares common understanding and agrees synthesis reflects their learning
When validation reveals weaknesses: return to data sources for low fidelity, refine statements for weak actionability, conduct targeted research for completeness gaps, abandon themes for insufficient robustness.
Coaching Patterns
Every major theme requires evidence from multiple research sources
Include all key user groups; silent stakeholders often reveal critical constraints
Maintain domain-specific nuances while identifying universal patterns
Do not force themes that evidence does not genuinely support
Quality Rules
Seven red flags signal synthesis failure: Single Source Dependency, Stakeholder Blind Spots, Pattern Forcing, Solution Bias, Jargon Overload, Scope Creep, and Premature Convergence
Effective synthesis demonstrates multi-source validation, complete stakeholder representation, actionable insights, robust patterns, and preserved context
Test themes with the question: would original research participants recognize themselves in this synthesis?
Synthesis Goals
Accomplish
Multi-source pattern recognition across different research data types
Stakeholder perspective integration across all key user groups
Environmental context preservation including domain-specific constraints
Solution-ready problem statements without prescribing solutions
Avoid
Theme forcing where evidence does not support connections
Single source over-reliance
Premature solution jumping before completing synthesis
Context loss in pursuit of generic patterns
Success Indicators
During Synthesis
All research inputs systematically analyzed without data loss
Themes supported by multiple independent data sources
Environmental and contextual factors preserved
Team members demonstrate shared understanding
After Synthesis
Unified problem statements with stakeholder consensus
Validated themes supported by comprehensive evidence
Solution focus areas identified without premature solution prescription
Team alignment on problem understanding and design direction
Input from Method 2
User interview findings with direct quotes and observations
Environmental constraint documentation
Workflow integration requirements
Unmet need patterns across user groups
Output to Method 4
Clear problem focus areas with validated themes for solution development
Design constraints from environmental and contextual factors
Stakeholder priorities and specific needs for solution targeting
Success criteria based on synthesized problem understanding
Artifacts
Method 3 artifacts at .copilot-tracking/dt/{project-slug}/method-03-synthesis/:
affinity-clusters.md: labeled clusters with representative evidence and tensionsinsight-statements.md: synthesized insights explaining why patterns matter, with supporting sources and design implicationsproblem-definition.md: unified problem framing with scope, stakeholders, constraints, and success signalshow-might-we-questions.md: HMW questions bridging from problem understanding to ideation, with related insights and constraints
Problem-to-Solution Space Transition
Method 3 sits at the boundary between Problem Space and Solution Space. This is the most critical transition. Moving to solutions without validated problem understanding produces solutions to the wrong problem.
Transition readiness signals:
Synthesis validation shows strength across all five dimensions with remaining gaps explicitly acknowledged
The team can articulate the discovered problem in terms that differ meaningfully from the original request
Multiple stakeholder perspectives are represented in the synthesis themes
Environmental and workflow constraints are documented, not just functional requirements
Next-step pathways:
Proceed to Method 4 when synthesis artifacts are complete and stable
Hand off to the RPI workflow through the Problem Space exit procedure in the
dt-rpi-integrationskill when the team wants a Researcher/Planner/Implementor to continue from validated problem understandingReturn to Methods 1-2 when synthesis exposes research gaps, conflicting narratives, or missing constraints
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.
Attribution and licence
- Published by
- Microsoft
- Publisher
- Microsoft (Official)
- Source
- microsoft/hve-core · .github/skills/design-thinking/dt-methods · commit f425e94
- Licence
- MIT License (redistribution allowed)
- LICENSE (added by AutoGineer from LICENSE)
- ATTRIBUTION.md (written by AutoGineer)
- Synced
AutoGineer is not affiliated with or endorsed by Microsoft.
Download ZIP
The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download dt-methods.zip