The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary functionality involves parsing and evaluating technology data provided by users, which could contain embedded instructions to influence the agent's behavior. Ingestion points: The scripts/format_detector.py module parses raw user input in Text, JSON, and YAML formats. Boundary markers: No delimiters or instructions are used to separate user data from agent directives. Capability inventory: The scripts/report_generator.py script includes the ability to write files to the local filesystem. Sanitization: Input data is not sanitized for potential LLM injection markers.
[COMMAND_EXECUTION]: The report_generator.py module exposes an export_to_file function that writes content to a path specified by an argument, which could be exploited to overwrite sensitive files if the path selection is manipulated.

tech-stack-evaluator