The Agent Skills Directory

Data Processing Workflow: The skill is designed to perform data transformation, reading from local inputs like mapped.json and user-provided CSV/JSON exports to create a structured evidence.json file. This is a standard and isolated data processing pattern.
Handling of External Data: The skill processes user-uploaded evidence indexes and exports. While ingesting external data is a potential surface for indirect prompt injection, the risk is minimal in this context because the skill's instructions focus on structured data extraction (date, area, person) rather than natural language generation or command execution.
Static Reference Usage: The skill relies on local reference files (clause-table.json and audit-rules.md) for its logic, ensuring that the evidence gathering process remains consistent with established ISO standards and internal rules.
Schema Standardization: The use of a JSON schema in contracts/mfg.audit-readiness.v1.json facilitates predictable data structures, which is a positive security practice for maintaining data integrity between different stages of an audit workflow.

evidence-gather