The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes several local Python utility scripts (sf-scan.py, sf-trace.py, sf-verify-doc.py) located in a scripts directory. These tools are used to scan project metadata, generate traceability matrices, and perform structural validation of the generated documents.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from Product Requirements Documents (prd.md) and the project filesystem to inform its generation process.
Ingestion points: The skill reads prd.md files from the feature directory and performs a project-wide scan via sf-scan.py to extract context.
Boundary markers: There are no explicit instructions or delimiters used when interpolating external PRD content to prevent the model from following malicious instructions embedded within those documents.
Capability inventory: The skill can execute shell commands via local Python scripts and has write access to the filesystem (creating and updating srs.md files).
Sanitization: While the skill uses a validation script (sf-verify-doc.py) and a manual checklist, these are designed to verify document completeness and compliance with IEEE standards rather than sanitizing or filtering malicious prompt injection patterns.

srs-generation