The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script named campaign-tracker.py with arguments like --brand {slug} and --action list-campaigns. This script is not included in the skill package and represents a dependency on an external executable.
[INDIRECT_PROMPT_INJECTION]: The skill processes multiple external files (brand profiles, messaging guidelines, and compliance rules) to generate its output, which could be exploited if those files contain malicious instructions. 1. Ingestion points: Reads files from ~/.claude-marketing/brands/{slug}/ and skills/context-engine/. 2. Boundary markers: No delimiters or safety instructions are defined to separate the brand data from the agent's core instructions. 3. Capability inventory: Subprocess execution via python campaign-tracker.py and extensive filesystem read access (SKILL.md). 4. Sanitization: The instructions do not describe any sanitization or validation of the brand profile content before it is used to influence the agent's response.

funnel-architect