The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script located at ${CLAUDE_PLUGIN_ROOT}/scripts/campaign-health-monitor.py. This script is used to calculate health scores, retrieve history, and generate reports. The execution uses variable interpolation for arguments such as brand slugs, campaign IDs, and timestamps.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for indirect instructions through external data sources.
Ingestion points: Reads brand profiles from ~/.claude-marketing/brands/{slug}/profile.json, agency SOPs from ~/.claude-marketing/sops/, and campaign metrics provided during the monitoring process.
Boundary markers: No explicit boundary markers or 'ignore' instructions are defined in the skill text for the ingested data.
Capability inventory: The skill has the capability to execute the campaign-health-monitor.py script via a subprocess.
Sanitization: No explicit sanitization or validation logic is described for the content of the JSON profiles or SOP files before they are processed by the agent.

autopilot-status