Product design skills

silver-research

Define research questions and methods, participant criteria, scripts, evidence handling, and sanitized observation plans. Use to plan product research or evaluation recruitment without implying the planned research was conducted.

Category
User research
Status
Experimental
Use when
  • plan ethical user research
  • define research questions and participant criteria
  • write a research plan without claiming it was conducted
  • plan evaluation or usability-test recruitment
  • define consent and data minimization for a study
Works on
  • a decision that needs research evidence and the question it must answer
  • optional existing product, problem-frame, or design-specification context
Produces
  • a research plan covering method, participant criteria, script, and consent/minimization/sanitation/retention handling, explicitly marked as planned rather than conducted
Requires
node
Works with
Claude Code, Codex, Cursor, OpenCode, Amp, Gemini CLI, GitHub Copilot, VS Code, Claude.ai
Tags
user-research, interviews, qualitative, quantitative
Source
thejparsenault/silver-design-framework by JP Arsenault · MIT
Related skills
user-research-cookiysurvey-designsilver-synthesize

Instructions from SKILL.md

Plan research

Workflow

  1. Define the decision the research must inform and the evidence needed.
  2. Choose a proportionate method, participant criteria, tasks, and script.
  3. Define consent, minimization, sanitation, retention, and evidence labeling.
  4. Mark the artifact as a plan until observed evidence is actually captured.

Run the guarded file operation through the CLI when durable outputs are ready:

.silver/bin/silver invoke --scaffold research .
.silver/bin/silver invoke research <request.json> .

The scaffold prefills timestamps, provenance, pinned context, required checks, and expected_integrity. Replace every silver-scaffold-placeholder before invoking; the CLI refuses a request that still contains one.

Done

  • Satisfy: question-method-aligned, participant-criteria-explicit, planned-not-conducted.
  • Evaluate quality: The plan can answer the declared question while minimizing sensitive data collection.
  • Emit a valid silver/skill-result/v2 record separating execution, acceptance, and downstream readiness.
  • Recommend follow-up skills; never start them automatically.

Boundaries

  • Never claim planned sessions, participants, or observations occurred.
  • Do not place sensitive raw participant data in the repository.

Supporting files

4 more files in this skill's folder. Markdown opens as a page, other text files open as source, and anything else downloads as it is.

Next to SKILL.md2
agents/1
scripts/1

On this page