The Agent Skills Directory

[SAFE]: No malicious code or security threats were identified. The skill's behavior is consistent with its stated purpose of document generation based on local context.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests and extracts context from an external file (CUSTOMER.md). This data is used to pre-fill recommendations and influence the generated content. Ingestion points: The skill reads CUSTOMER.md in Phase 0. Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt logic. Capability inventory: The skill is restricted to file reading, user interaction via AskUserQuestion, and local file writing of BRAND_GUIDELINES.md. Sanitization: No specific sanitization or validation of the input file content is performed.

define-brand-guidelines