deep-topic-research

Auditable topic research requires deterministic logging of web searches and gathered sources. This skill validates evidence and generates structured artifacts like evidence.json and citation-report.json, enabling complete traceability from final findings back to original sources.

Deterministic deep-topic research uses fixed scripts to log searches and assemble validated evidence. You need it when findings must be reproducible and fully auditable, such as verifying product claims or analyzing codebases with strict source validation requirements.

Yes, repository analysis and deterministic deep-topic research require Python 3.12+ for running scripts and network access for web search and fetch operations. These dependencies ensure sources are gathered and validated in an auditable manner.

Initialize a research run and the workflow logs searches, gathers sources, and validates evidence. It outputs structured artifacts including domain-plan.json, analysis.json, validation.json, and a final report.md summarizing the auditable findings.

Evidence validation during codebase research is best handled by deterministic scripts that log every search query and fetched source. This produces an evidence.json file and a citation-report.json, ensuring all repository analysis findings are auditable and verifiable.

Structured artifacts like requirements.json and domain-plan.json are needed to maintain deterministic traceability. They capture the research plan, validated evidence, and analysis separately, ensuring the final report.md is fully auditable and reproducible.