The Agent Skills Directory

[SAFE]: The skill consists primarily of educational content and templates for business insurance. No malicious code, prompt injection, or data exfiltration patterns were identified.
[COMMAND_EXECUTION]: The skill includes a Python script scripts/check-output.py intended to validate deliverables. The script performs simple string matching and length checks on a provided file path. It does not execute external commands or interact with the network.
[INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface via scripts/check-output.py, which reads the contents of a file provided as a command-line argument.
Ingestion points: scripts/check-output.py reads user-generated markdown files.
Boundary markers: None present in the ingestion script.
Capability inventory: The script is limited to reading a file and printing results to standard output; it has no network, file-write, or execution capabilities.
Sanitization: The script converts text to lowercase for comparison but does not perform complex sanitization. However, due to the limited capabilities, this does not pose a security risk.

business-insurance