The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external brand profiles and guidelines (e.g., ~/.claude-marketing/brands/{slug}/profile.json, _manifest.json). This presents a potential surface for indirect prompt injection if those files are provided by an untrusted source. 1. Ingestion points: Files located in ~/.claude-marketing/brands/ and skills/context-engine/compliance-rules.md. 2. Boundary markers: None specified in the instructions. 3. Capability inventory: No subprocess calls, network operations, or dynamic code execution identified in the skill logic. 4. Sanitization: No explicit sanitization or validation of the input file content is mentioned.
[SAFE]: The skill is designed for planning purposes only, focusing solely on generating structured media plan documentation without accessing sensitive system credentials or executing remote code.

media-plan