The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data access detected across the analyzed files.- [PROMPT_INJECTION]: The skill uses natural instructional language to define its persona and mission. No attempts to bypass safety filters, extract system prompts, or override core agent instructions were found.- [DATA_EXFILTRATION]: The skill does not perform network operations or access sensitive local files (e.g., credentials, SSH keys, or environment variables).- [REMOTE_CODE_EXECUTION]: No external dependencies are fetched or executed. The included TypeScript snippet in SKILL.md is a static template generator that returns string literals and does not invoke external processes or perform dynamic evaluation.- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied 'creative briefs' (Phase 1). While this is an ingestion point for untrusted data, the skill lacks capabilities (such as shell execution or file writing) that would make it exploitable via indirect injection. The risk is considered negligible.

agency-inclusive-visuals-specialist