doc-coauthoring

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates reading content from external sources such as Google Drive, SharePoint, Slack, and Microsoft Teams through integrations.
  • Evidence: The workflow instructs the agent to use available integrations to fetch documents and chat history when provided by the user to gather context for the drafting process.
  • [COMMAND_EXECUTION]: The skill uses file manipulation tools to create and refine documentation locally.
  • Evidence: The instructions call for using str_replace or the Edit tool to build markdown files and perform iterative updates based on user feedback.
  • [PROMPT_INJECTION]: The skill processes untrusted external data during the context-gathering stage, creating a surface for potential indirect prompt injection.
  • Ingestion points: The agent ingests context from shared documents, links, and team communication threads (file: SKILL.md).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified for the ingested content.
  • Capability inventory: The agent possesses file-writing capabilities through the use of editing tools to draft and update documentation (file: SKILL.md).
  • Sanitization: No sanitization or validation of the external content is implemented in the workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 06:46 PM