silver-synthesize
Convert sanitized evidence, feedback, analytics, briefs, and labeled assumptions into findings, problem frames, opportunities, contradictions, and open questions with provenance. Use after research, feedback, evaluation, analytics, or a seed brief.
- Category
- User research
- Status
- Experimental
- Use when
- synthesize research evidence into findings
- turn feedback and analytics into a problem frame
- cluster evidence without losing contradictions
- separate assumptions from findings
- produce traceable findings with provenance
- Works on
- sanitized evidence with provenance: observations, analytics, supplied facts, labeled assumptions
- optional prior evaluation or map artifacts
- Produces
- findings and a problem frame that keep observation, interpretation, confidence, contradiction, and open questions distinguishable and traceable to their sources
- Requires
- node
- Works with
- Claude Code, Codex, Cursor, OpenCode, Amp, Gemini CLI, GitHub Copilot, VS Code, Claude.ai
- Tags
- synthesis, thematic-analysis, affinity-mapping, insights
- Source
- thejparsenault/silver-design-framework by JP Arsenault · MIT
- Related skills
- silver-researchuser-research-cookiysilver-ideate
Instructions from SKILL.md
Synthesize evidence
Workflow
- Confirm every source is sanitized and revision-addressable.
- Separate observations, analytics, supplied facts, and assumptions.
- Cluster evidence without erasing contradictions or minority signals.
- Produce traceable findings and an actionable problem frame.
Run the guarded file operation through the CLI when durable outputs are ready:
.silver/bin/silver invoke --scaffold synthesize .
.silver/bin/silver invoke synthesize <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: sources-pinned, assumptions-labeled, contradictions-retained.
- Evaluate quality: Findings separate observation, interpretation, confidence, contradiction, and open question.
- Emit a valid
silver/skill-result/v2record separating execution, acceptance, and downstream readiness. - Recommend follow-up skills; never start them automatically.
Boundaries
- Do not fabricate evidence or silently upgrade assumptions into findings.
- Do not rewrite upstream evidence while synthesizing it.
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.