doc-importer
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from external files (PDF, DOCX, etc.) and URLs, creating an attack surface where malicious instructions embedded in those documents could influence the agent's behavior.
- Ingestion points: The workflow in
SKILL.md(Step 1 and Step 2) specifies reading content from local file paths and external URLs provided by the user. - Boundary markers: The skill includes a mandatory sanitization step (Step 4) that requires wrapping external content in boundary markers to help the agent distinguish between instructions and data.
- Capability inventory: The skill utilizes file read/write capabilities and network access for URL retrieval, and it can hand off processed content to other skills like
scribe:doc-generator. - Sanitization: Step 4 explicitly instructs the agent to strip system/instruction tags and truncate large sections, which serves as a defensive measure against payload execution.
Audit Metadata