makepad-font

Pass

Audited by Gen Agent Trust Hub on Apr 26, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill uses authoritative language such as 'Claude MUST' and 'IMPORTANT' to define a specific workflow for checking documentation completeness. These are meta-instructions for the agent's task-specific behavior and do not attempt to bypass underlying safety filters or extract system prompts.
  • [PROMPT_INJECTION]: The skill exhibits a surface area for indirect prompt injection because it instructs the agent to read and incorporate content from local reference files (e.g., ./references/font-system.md) into its responses.
  • Ingestion points: Reading local markdown files via the agent's file system tools.
  • Boundary markers: None identified; the agent is not instructed to treat the content of these files as untrusted.
  • Capability inventory: The skill body contains no dangerous capabilities such as arbitrary command execution, network writes, or file modification tools.
  • Sanitization: No sanitization of the reference content is performed before interpolation into the agent's context.
  • [COMMAND_EXECUTION]: The skill text mentions a command (/sync-crate-skills makepad --force) that the agent should suggest to the user if local documentation is missing. This is a user-facing recommendation for environment synchronization and does not involve the agent executing the command itself.
  • [SAFE]: The external URLs provided in the documentation (e.g., crates.io and example.com) are used for informational purposes and example links, representing no security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 26, 2026, 06:59 AM