evidence-extraction

Evidence extraction standardizes findings from artifacts into a memory format to ensure traceability and governance. It reads artifacts from claude-memory contexts, assigns unique Finding IDs, and writes normalized results to memory.md.

Maintaining a technology inventory involves extracting and recording tools and platforms mentioned in artifacts during evidence extraction. This builds an inventory within the memory context to support Stage 9 output and auditing processes.

Regulatory context detection scans artifacts for explicit or inferred compliance requirements during evidence extraction. It records this context in memory to enable compliant workflow execution and ensure governance traceability.

Yes, evidence extraction assigns confidence levels to findings alongside unique Finding IDs. It normalizes extracted data and records sources without orchestrating other agents, ensuring structured traceability within the memory.md file.

Evidence extraction operates on artifacts registered in claude-memory contexts during Stage 1. It classifies artifact types, extracts findings, and writes results to claude-memory/memory.md to support governance and traceability.

Evidence extraction does not orchestrate other agents and operates solely on artifacts within claude-memory contexts. It focuses on normalizing findings and building the technology inventory without performing downstream workflow actions.