The Agent Skills Directory

[SAFE]: The skill documentation and scripts focus exclusively on business intelligence and competitor analysis workflows. All logic is transparent and follows standard data processing patterns.
[COMMAND_EXECUTION]: The included Python scripts (battle_card_generator.py, competitor_scorer.py, and feature_matrix_builder.py) process user-provided JSON files and output formatted text or Markdown. They do not execute arbitrary shell commands or use dangerous functions such as eval() or exec().
[EXTERNAL_DOWNLOADS]: The skill does not perform any network operations or download external code. It references well-known business intelligence services (e.g., Ahrefs, G2, Capterra) as suggested manual data sources for the user.
[DATA_EXFILTRATION]: No patterns for data exfiltration were found. The scripts operate locally on data provided as command-line arguments and print results to standard output.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external competitor data (such as user reviews and website content). While this creates an inherent surface for indirect prompt injection if the data is processed by an LLM, the provided scripts perform deterministic formatting and do not involve sensitive agent capabilities or automated feedback loops. This is a standard risk for analysis tools.

competitive-teardown