doc-coauthoring
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows a legitimate and structured process for collaborative document creation and review.
- [COMMAND_EXECUTION]: The skill utilizes standard platform tools such as create_file and str_replace to manage document versions and artifacts, which is expected behavior for an agent assisting with writing tasks.
- [DATA_EXFILTRATION]: The skill requests access to organizational context and documents via official platform integrations to inform its drafting process. No unauthorized exfiltration patterns or external network requests to untrusted domains were identified.
- [PROMPT_INJECTION]: The skill has an ingestion surface for untrusted data from shared documents and messaging channels. Boundary markers for this data are absent, and the agent has the capability to modify files using create_file and str_replace. However, the workflow includes human-in-the-loop curation and a 'Reader Testing' verification phase to validate outputs before completion.
Audit Metadata