@tank/character-creator

Pass

Audited by Gen Agent Trust Hub on May 31, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses official APIs from well-known technology services (OpenAI and fal.ai) to perform its core character generation tasks.
  • [SAFE]: All referenced external dependencies, such as 'sharp', 'Pillow', and the official OpenAI and fal.ai clients, are reputable and widely used in the industry.
  • [SAFE]: The skill requested permissions for 'subprocess' and 'filesystem' write access are justified by the stated need to install development dependencies and manage local image assets during the assembly process.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection (Category 8) due to the interpolation of user-supplied character descriptions into generative AI prompts.
  • Ingestion points: Character details gathered during the discovery interview phase defined in 'references/character-discovery.md'.
  • Boundary markers: The pipeline utilizes clear template placeholders (e.g., [CHARACTER_DNA_BLOCK]) to structure user-provided data within system prompts.
  • Capability inventory: The skill has the capability to write to the local filesystem (saving character JSON and images) and execute subprocesses (e.g., installing required npm packages).
  • Sanitization: There is no explicit evidence of sanitization or escaping of the user-provided text before it is inserted into the prompt templates.
Audit Metadata
Risk Level
SAFE
Analyzed
May 31, 2026, 11:04 PM
Security Audit — agent-trust-hub — @tank/character-creator