The Agent Skills Directory

[SAFE]: Analysis of the skill instructions and associated scripts indicates a focused intent on marketing data processing. The skill does not perform suspicious network operations, access sensitive system files, or attempt to bypass security controls.
[INDIRECT_PROMPT_INJECTION]: The skill's workflow involves reading and processing external data files, which presents a surface for indirect prompt injection if those files contain malicious instructions.
Ingestion points: Scripts channel_mix_optimizer.py, cohort_analyzer.py, and marketing_forecast_generator.py load data from user-supplied JSON files.
Boundary markers: The provided instructions lack specific directives for the agent to use delimiters or ignore potentially adversarial instructions embedded within the data sources.
Capability inventory: The skill is limited to data computation and stdout reporting; it does not have capabilities for network communication, file system modification, or dynamic code execution.
Sanitization: Input data is parsed using standard library JSON functions without additional sanitization or instruction-filtering logic.

marketing-analyst