character-style-consistency-by-1enbgh4
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references external URLs from goodcase.ai, media.goodcase.ai, and x.com in the references/cases.md file. These links are provided to allow the agent to inspect visual evidence and style documentation for the character consistency workflow. These references are used neutrally for style analysis and source attribution.
- [INDIRECT_PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection as it ingests user-supplied subjects, audience details, and brand assets to generate final output prompts. According to the evidence chain, ingestion occurs through the input parameters defined in SKILL.md. The skill mitigates this by implementing clear boundary markers in the form of a 'Preserve / Replace / Avoid' reference contract, ensuring that core style traits remain isolated from untrusted user content. The capability inventory is restricted to generating text-based prompts with no evidence of file-write, network exfiltration of sensitive data, or command execution capabilities.
Audit Metadata