addon-report-synthesis-audit
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect prompt injection risk. The skill ingests untrusted data from documents which could contain instructions intended to manipulate the final report summary. * Ingestion points: The workflow loads
documents,extracted_clauses, andvalidationsfrom external storage (SKILL.md). * Boundary markers: Absent. No specific instructions are provided to the agent to ignore instructions embedded within the ingested data. * Capability inventory: The skill writes to Postgres databases (reports.summary_jsonb) and S3/MinIO object storage (documents/reports/) (SKILL.md). * Sanitization: Absent. No filtering or escaping mechanisms are mentioned for the processed data.
Audit Metadata