product-quality-perspective
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of Markdown-based prompts and YAML configuration files used to guide an AI agent in performing product quality assessments (e.g., code reviews, test strategies). Analysis of the instructions and workflow revealed no malicious patterns.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted project materials such as requirements, code diffs, and test reports. While this presents an injection surface, the skill implements strict boundaries, including a 'Report Contract' and a 'Pre-delivery Checklist' to distinguish facts from inferences and prevent the invention of data. The skill lacks dangerous capabilities (network access, command execution, or file writing), making the risk minimal.
- Ingestion points: Project materials provided as input to the specific quality stages (e.g., PR/diffs, requirements documents).
- Boundary markers: Explicit instructions in
SKILL.mdandprompts/*.mdto separate facts, evidence, and inferences. - Capability inventory: No tool use or system-level capabilities are requested or utilized by the skill.
- Sanitization: Relies on structured output reporting rather than explicit input sanitization.
Audit Metadata