pencil-mcp-set-variables

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is properly scoped to design tasks and includes proactive safeguards. It requires the agent to verify user intent specifically for the 'Pencil' tool, preventing misuse during unrelated tasks.
  • [PROMPT_INJECTION]: Analysis of the instructions shows no attempts to override agent behavior, bypass safety protocols, or extract sensitive internal prompts.
  • [DATA_EXPOSURE]: While the skill accepts file paths and variable data, these are standard inputs for the associated tool. No patterns of credential harvesting or sensitive file exfiltration were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 10:48 AM