quality-manager-qmr
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize qualitative data from external sources, including customer feedback, complaint narratives, and external audit results, which could contain adversarial instructions.
- Ingestion points: Data enters the agent's context through the "Management Review Input Summary" and "Required Inputs" workflows defined in
SKILL.mdandreferences/management-review-guide.md. - Boundary markers: The provided templates for data collection do not utilize strict delimiters or specific system instructions to differentiate between data and instructions.
- Capability inventory: The skill includes Python scripts (
scripts/management_review_tracker.pyandscripts/quality_effectiveness_monitor.py) that perform file read operations and data analysis to generate executive reports. - Sanitization: There is no evidence of text sanitization or filtering for inputs processed by the agent or the analysis scripts.
Audit Metadata