client-package-notion
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes contents from strategy documents, campaign drafts, and lead lists stored in the local client folder. Because this data may originate from external sources, it presents an attack surface where embedded instructions (e.g., in a CSV field or Markdown file) could attempt to influence the agent's behavior during the packaging process.
- Ingestion points: Files within the
clients/<client_name>/directory, including.mdand.csvassets. - Boundary markers: The skill does not provide clear delimiters or instructions to the agent to distinguish between its own system instructions and the data being processed from the client files.
- Capability inventory: The skill has permissions to read local files and write data to external cloud services (Notion and Google Sheets).
- Sanitization: While the skill involves formatting text for Notion Markdown compatibility, it lacks semantic sanitization to prevent the agent from obeying instructions hidden within the data.
Audit Metadata