doc-coauthoring

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches context from official service integrations (Slack, Google Drive) and references a trusted repository (github.com/anthropics) for project attribution. These operations are neutral and consistent with a documentation workflow.
  • [COMMAND_EXECUTION]: The skill uses standard file manipulation tools, including create_file and str_replace, to build and refine document sections based on user feedback.
  • [INDIRECT_PROMPT_INJECTION]: The skill manages a data ingestion surface where external content is processed. 1. Ingestion points: User info-dumps, external links, and team messaging channels (SKILL.md). 2. Boundary markers: None defined. 3. Capability inventory: Writing files (create_file), editing text (str_replace), and spawning sub-agents for testing (SKILL.md). 4. Sanitization: Not specified. While this presents an injection surface, it is consistent with the skill's primary purpose and contains no malicious patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 08:28 AM
Security Audit — agent-trust-hub — doc-coauthoring