The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script audience-simulator.py using shell commands. User-supplied inputs, such as product descriptions and segment definitions, are passed as arguments to this script, creating a potential vector for shell command injection if the inputs contain metacharacters (e.g., ;, &, |).
[DYNAMIC_EXECUTION]: The process involves dynamic construction of command-line strings for the audience-simulator.py tool. The skill interpolates variables like {slug}, {name}, {id}, and {price-points} directly into the execution string.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the processing of untrusted data.
Ingestion points: User-provided product descriptions, segment definitions, and competitive pricing context (SKILL.md).
Boundary markers: None identified; user inputs are directly interpolated into the script arguments.
Capability inventory: The skill has the capability to execute shell commands via Python, read brand profiles, guidelines, and CRM-grounded audience data from the local file system (e.g., ~/.claude-marketing/).
Sanitization: No sanitization, escaping, or validation of user-provided content is specified before it is passed to the underlying execution tool.

pricing-test