The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a Python script (campaign-tracker.py) located in the plugin's root directory to list campaign history. This is a legitimate requirement for marketing auditing.
[DATA_EXFILTRATION]: The skill accesses brand profiles and messaging guidelines stored in the user's home directory at ~/.claude-marketing/. While this involves file system access, it is restricted to the skill's own application data and is necessary for providing brand-consistent output.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest customer research data, which creates a surface for indirect prompt injection. 1. Ingestion points: CRM data, analytics, and survey results are processed as described in the 'Discovery & Data Collection' section of SKILL.md. 2. Boundary markers: The instructions do not define specific delimiters or warnings to isolate untrusted data. 3. Capability inventory: The skill can read brand context and execute local campaign tracking scripts. 4. Sanitization: No explicit validation or filtering logic for ingested data is provided.

audience-intelligence