doc-restructure-live
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes the full content of untrusted documents, which could contain instructions meant to hijack the agent's behavior.
- Ingestion points: The skill reads content from Google Drive/Docs via a connector as described in the 'Execution' section of
SKILL.md. - Boundary markers: There are no explicit markers or instructions provided to the agent to disregard instructions that might be embedded within the document text itself.
- Capability inventory: The skill includes the capability to write and create new files/documents within the user's Google Drive environment (
SKILL.md, Execution Step 4). - Sanitization: The skill does not implement sanitization or filtering to prevent the interpolation of malicious instructions from the source document into the agent's output context.
Audit Metadata