The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes descriptive data to generate reports, creating a theoretical attack surface for indirect prompt injection.
Ingestion points: The assess_risk function in scripts/process.py accepts breach_description, assessor_name, and breach_reference as inputs.
Boundary markers: No explicit delimiters or instructions are used to separate user-provided data from report templates.
Capability inventory: The script is limited to string processing and console output. It does not perform network operations, file system access, or command execution.
Sanitization: No validation or escaping is applied to the input strings.

breach-risk-assessment