The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script with arguments derived from user input. Evidence: python "${CLAUDE_PLUGIN_ROOT}/scripts/setup.py" --create-brand "[brand name]". Risk: If the user-provided brand name contains shell metacharacters and is not properly escaped by the execution environment or the script, it could allow for arbitrary command execution on the host system.
[INDIRECT_PROMPT_INJECTION]: The skill collects a wide range of untrusted user data—including brand voice descriptions, identity statements, and sample content—which is stored and subsequently used by all other skills in the ecosystem. 1. Ingestion points: User responses to questions about brand identity, mission, voice dimensions, and content samples (SKILL.md). 2. Boundary markers: The skill does not define specific delimiters or instructions to prevent the agent from obeying instructions embedded within the brand profile data. 3. Capability inventory: The skill set includes the ability to execute Python scripts and modify local configuration files. 4. Sanitization: No sanitization, validation, or filtering of the ingested user content is documented.

brand-setup