The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted input data, creating a potential surface for indirect prompt injection.\n- Ingestion points: Data is read from a file path provided via the --input argument in run.py.\n- Boundary markers: The implementation lacks delimiters or instructions to ignore embedded commands within the input data.\n- Capability inventory: The skill's logic in run.py is limited to reading and writing local files; no network or dynamic code execution capabilities are present.\n- Sanitization: There is no evidence of input validation or sanitization before the data is processed by the reference implementation.

billing-webhook-analyzer