writing
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by design.
- Ingestion points: In
DOC MODE(Stage 1: Context Gathering), the instructions explicitly encourage using integrations such as Google Drive and Slack to pull context directly into the agent's workspace. - Boundary markers: There are no specified delimiters or instructions to the agent to ignore or isolate instructions that might be embedded within the data retrieved from external integrations.
- Capability inventory: The skill (specifically in
PLAN MODE) authorizes the agent to create and modify files (e.g., implementation plans and source code) and provides templates for executing shell commands such aspytestandgit commit. - Sanitization: No sanitization, validation, or escaping of the external content is mentioned before the agent processes or interpolates it into the documentation workflow.
Audit Metadata