Sprint 4: Canonical Model Completion

[Goose-Flight]

Sprint 4: Canonical Model Completion

Goal

Make the data model truly forensic-ready — evidence-linked findings, hypotheses, signal quality, and full provenance linkage.

Deliverables

• SignalQuality model
• EvidenceReference model (links a finding back to evidence item + stream + time range)
• Expanded Finding model (finding_id, plugin_id, evidence_references[], confidence, supporting/contradicting metrics)
• Hypothesis model (supporting_findings[], contradicting_findings[], confidence, unresolved_questions[])
• Provenance linkage across parsed and analyzed artifacts
• Serialization support for exports
• Plugins emit findings with evidence references (at least evidence_id + stream name)
• findings.json enriched with full forensic fields

Acceptance criteria

• Findings can cite evidence clearly (evidence_id, stream, time range)
• Hypotheses can link supporting and contradicting findings
• Parsed and analyzed artifacts are traceable back to evidence
• All models serialize/deserialize cleanly

[USTungsten]

Sprint 4 complete. 433 tests passing. Canonical forensic model delivered: ForensicFinding, EvidenceReference, SignalQuality, Hypothesis models; lifting layer; versioned findings/hypotheses bundles; Hypotheses tab in GUI. Pushed as commit 2583262.

[USTungsten]

Sprint 4 complete. 433 tests passing. Canonical forensic model delivered: ForensicFinding, EvidenceReference, SignalQuality, Hypothesis models; lifting layer; versioned findings/hypotheses bundles; Hypotheses tab in GUI. Pushed as commit 2583262.