doc-type-routing
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides a decision-making scaffold to categorize documents based on user input, which presents a surface for indirect prompt injection where malicious instructions embedded in a user's document description could attempt to influence agent behavior.
- Ingestion points: User descriptions of writing tasks (purpose, direction, and response requirements) in SKILL.md.
- Boundary markers: Absent; there are no specific delimiters or instructions defined to isolate or ignore potential commands within user-provided descriptions.
- Capability inventory: The skill references local configuration files and a Python script (
src/scripts/check_sections.py) for subsequent validation steps. - Sanitization: No sanitization, filtering, or validation of the input text is described in the provided instructions.
Audit Metadata