document
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted documentation files (e.g., requirements, specifications, and changelogs) which creates a surface for indirect prompt injection.
- Ingestion points: The skill reads
PROJECT-DOCUMENTS.md,CHANGELOG.md, and capability-specific markdown files from the repository. - Boundary markers: The instructions lack guidance for the agent to use delimiters or to treat instructions found within the processed files as data rather than commands.
- Capability inventory: The agent environment typically includes file-writing capabilities and the ability to call other skills, which could be exploited through injection.
- Sanitization: No sanitization or validation of the content within the documentation files is performed before the agent processes them.
Audit Metadata