The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's instructions in SKILL.md direct the agent to execute a local Python validator script (scripts/validate_sesf.py) to verify the structural integrity of generated specifications. This is a functional utility for the skill's primary purpose and does not involve remote code execution or execution of untrusted external content.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes user-provided inputs (user_request, domain) to generate specification documents. While a user could attempt to inject instructions, the skill's output is restricted to text generation and does not grant the agent high-privilege capabilities that would lead to a high-severity finding.
Ingestion points: The user_request and domain parameters in the SKILL.md input section.
Boundary markers: The skill uses an XML envelope structure (, , etc.) which provides clear structural boundaries for generated content, though it lacks explicit "ignore embedded instructions" warnings for the user-provided input itself.
Capability inventory: Execution of the included local validation script (scripts/validate_sesf.py); file system read access for validation tasks.
Sanitization: The validate_sesf.py script acts as a structural sanitizer, ensuring the generated output conforms to the HSF schema and flagging non-compliant structures.

hsf