The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes structured input from external files, which creates an attack surface for indirect prompt injection where data could contain instructions meant to override agent behavior. \n- Ingestion points: run.py reads content from a file path specified by the user in the --input argument. \n- Boundary markers: The skill does not implement delimiters or safety instructions to separate data from instructions within the input file. \n- Capability inventory: run.py performs file system operations including reading files (read_text), writing files (write_text), and creating directories (mkdir). \n- Sanitization: No content validation or sanitization is performed on the input data before it is processed by the run function.

sales-usage-analyzer