The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes an internal Python script using python "${CLAUDE_PLUGIN_ROOT}/scripts/growth-loop-modeler.py". This script is used for growth loop detection, bottleneck analysis, and forward projections based on provided business metrics.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from business metrics and brand profiles (~/.claude-marketing/brands/{slug}/profile.json) to perform its analysis.
Ingestion points: Reads brand profiles and business metrics (acquisition, retention, revenue) provided by the user.
Boundary markers: No specific delimiters or instruction-ignore markers are mentioned for the data ingestion phase.
Capability inventory: The skill is restricted to reading local brand files and running the specific growth-loop-modeler.py script. It does not have access to the network or system-wide write capabilities.
Sanitization: The instructions do not specify sanitization for the brand profile data, though the primary consumer is a mathematical modeling script.
[METADATA_POISONING]: The skill's metadata and description accurately reflect its core functionality of growth loop analysis and recommendation generation, with no deceptive or malicious instructions found in the triggers or author fields.

loop-detect