The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute local Python scripts (scripts/profile.py and various scripts in scripts/api/) for managing business context and verifying API connections across Meta, Google, and TikTok platforms.
[INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests and stores user-supplied business details and account identifiers.
Ingestion points: User responses collected via the AskUserQuestion tool throughout the onboarding flow in SKILL.md (Phases B, C, and D).
Boundary markers: None explicitly defined for input delimiters, though the flow is strictly controlled by the agent.
Capability inventory: Bash (executing Python scripts), Write (modifying ~/.claude-ads/profile.json), and Read capabilities.
Sanitization: The skill mitigates risks by mapping user selections to specific string constants (e.g., industry and goal types) and coercing budget inputs to numerical values before passing them to the configuration script.

ads-start