doc-coauthoring

Pass

Audited by Gen Agent Trust Hub on Nov 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted data from multiple external sources, creating a surface for indirect prompt injection attacks where malicious instructions hidden in documents or chat history could influence the agent.
  • Ingestion points: The workflow explicitly encourages pulling context from "team channels or threads" (Slack, Teams), "shared documents" (Google Drive, SharePoint), and external "MCP servers".
  • Capability inventory: The agent has capabilities to read external data and create/modify files (create_file, str_replace).
  • Boundary markers: The skill does not define specific delimiters or warnings to the model to ignore instructions embedded in the external content it fetches.
  • Sanitization: There are no explicit instructions for the agent to sanitize or ignore potentially malicious control sequences within the gathered context.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 7, 2026, 02:05 AM