pencil-mcp-get-style-guide

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it retrieves style guide metadata from an external tool and incorporates it into subsequent agent actions.
  • Ingestion points: Metadata enters the agent context from the output of the get_style_guide tool call (SKILL.md).
  • Boundary markers: The instructions lack specific delimiters or directions for the agent to ignore any embedded instructions that might be present in the fetched metadata.
  • Capability inventory: The skill specifies using the retrieved metadata in sensitive downstream tool calls such as batch_design or set_variables (SKILL.md).
  • Sanitization: No sanitization, filtering, or validation steps are defined to ensure the style guide content does not contain malicious prompt overrides.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 06:12 PM
Security Audit — agent-trust-hub — pencil-mcp-get-style-guide