project-delivery-perspective
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes project-delivery and quality facts supplied by users, which constitutes a data ingestion surface. It mitigates risks by requiring source attribution for every fact and providing explicit instructions in
SKILL.mdand the prompts to separate delivery tracking from quality data. The agent is strictly forbidden from changing defect or execution status even if requested by a stakeholder, a behavior verified in theedge-quality-fact-override.yamlevaluation case. - [DATA_EXPOSURE_AND_EXFILTRATION]: No access to sensitive file system paths (such as
.ssh,.aws, or.env) was found. The skill does not perform any network operations to external domains. - [REMOTE_CODE_EXECUTION]: The skill does not download or execute remote scripts, nor does it define any third-party package dependencies in standard package management files. All logic is implemented via Markdown prompts.
Audit Metadata