character-designer

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a template engine that ingests external data to generate character designs and prompts, creating a surface where untrusted data could influence agent outputs.
  • Ingestion points: Processes character descriptions provided by the user and reads brand guidelines from project-context.md (as seen in the 'Instructions' section).
  • Boundary markers: The templates do not utilize specific delimiters or isolation markers to distinguish between instructions and ingested data when generating prompts.
  • Capability inventory: The skill defines a file-writing workflow to save markdown files into the ./output/characters/ directory. It does not invoke shell commands, network requests, or dynamic code execution.
  • Sanitization: There are no defined sanitization or validation steps for the input data before it is interpolated into the final character archives or tokens.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 01:59 PM
Security Audit — agent-trust-hub — character-designer