The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run a local Python script geo-tracker.py using subcommands audit-visibility and diff. This script is not included in the provided skill files, representing an unverified dependency execution.
[DATA_EXFILTRATION]: The skill accesses sensitive brand configuration and profile data located at ~/.claude-marketing/brands/. While these paths are application-specific, the ingestion of local file content into the agent's context poses a risk of exposure.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources.
Ingestion points: Full response text is retrieved from external AI engines including ChatGPT, Perplexity, Gemini, AI Overviews, and Copilot (Process Step 3).
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified when processing these external responses.
Capability inventory: The agent has the capability to execute shell commands (geo-tracker.py) and read/write files in the ~/.claude-marketing/ directory.
Sanitization: There is no evidence of sanitization or filtering of the external AI responses before they are analyzed for narrative alignment.

geo-monitor