conduct-retrospective
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection surface. The skill processes external files without sanitization or boundary markers, creating a risk that malicious instructions within those files could be obeyed by the agent. * Ingestion points: Step 1 reads files matching STATUS-REPORT-*.md, SPRINT-PLAN.md, and BACKLOG.md. * Boundary markers: No delimiters or instructions to ignore embedded content are present in the procedure. * Capability inventory: The skill utilizes Read, Write, Edit, Bash, Grep, and Glob tools across all scripts. * Sanitization: No content validation or escaping is performed on the ingested data.
Audit Metadata