google-docs
Pass
Audited by Gen Agent Trust Hub on Apr 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Downloads the
@membranehq/clipackage from the npm registry, which is the official tool for interacting with the author's integration platform. - [COMMAND_EXECUTION]: Uses the Membrane CLI to perform authenticated operations on Google Docs, including searching for, creating, and running document actions.
- [PROMPT_INJECTION]: The skill interacts with external document data, creating a potential surface for indirect prompt injection.
- Ingestion points: Document content retrieved from Google Docs via the
get-documentaction in SKILL.md. - Boundary markers: Absent from the skill instructions.
- Capability inventory: Commands such as
batch-update-documentanddelete-contentallow the agent to modify or remove data. - Sanitization: No specific sanitization or filtering of document content is implemented.
Audit Metadata