The Agent Skills Directory

[SAFE]: The skill consists of instructional markdown and reference files that define a rigorous peer-review and validation process. It enforces best practices in requirements engineering, such as deterministic test oracles and bidirectional traceability.
[COMMAND_EXECUTION]: The skill refers to the use of a local Python script, scripts/validate_sdd_phase_boundaries.py, for structural validation of phase boundaries. This execution is documented as a standard development workflow tool for ensuring repository consistency and does not involve downloading remote code or executing untrusted inputs from the internet.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external requirements artifacts (e.g., requirements_analysis_report.md). While this creates a theoretical surface for indirect prompt injection, the skill includes explicit defensive instructions such as "Plausible generated prose is not validation evidence" and requires every pass to cite specific quality attribute evidence, mitigating the risk of the agent being misled by instructions hidden in the processed documents.
Ingestion points: Files located in projects/<ProjectName>/<phase>/<document>/.
Boundary markers: None explicitly defined for input data delimiters.
Capability inventory: File system read access, file system write access for report generation, and execution of a specific local Python script.
Sanitization: Not explicitly mentioned, but the workflow requires human-verifiable evidence for every validation outcome.

07-requirements-validation