The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, specifically compliance policies and evidence documentation.
Ingestion points: SKILL.md (Workflow steps involving mapping evidence and identifying gaps).
Capability inventory: No dangerous capabilities such as subprocess calls, network operations, or file system writes were identified.
Boundary markers: The instructions do not define specific delimiters for the ingested data.
Sanitization: No specific sanitization methods are mentioned for the provided input files.
[NO_CODE]: The skill consists entirely of natural language instructions and metadata. It does not contain any executable scripts, binary files, or external package dependencies.

compliance-readiness