doc-coauthoring

Pass

Audited by Gen Agent Trust Hub on Jun 27, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill manages the documentation process through the use of create_file and str_replace to build and refine document artifacts locally.
  • [DATA_EXPOSURE]: The skill facilitates access to external context by encouraging the use of platform connectors and MCP integrations for services like Slack, Microsoft Teams, and Google Drive.
  • [PROMPT_INJECTION]: The skill processes untrusted user-provided content and external message data to draft documentation, which serves as a surface for indirect prompt injection. Ingestion points: External content from messaging apps and shared documents (SKILL.md). Boundary markers: Absent; the instructions do not implement specific delimiters for ingested text. Capability inventory: Includes file creation and modification tools. Sanitization: None; external data is incorporated into the drafting process based on user curation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 27, 2026, 11:03 AM
Security Audit — agent-trust-hub — doc-coauthoring