The Agent Skills Directory

[SAFE]: The skill acts as a procedural guide and does not contain any functional code, shell commands, or network operations. It focuses on the creation and management of documentation files related to system validation.
[INDIRECT_PROMPT_INJECTION]: The skill processes external inputs such as user needs and stakeholder statements to generate validation plans. While this represents an ingestion surface for potentially untrusted data, the agent lacks high-risk capabilities (like network exfiltration or system commands) to execute malicious instructions embedded in that data. * Ingestion points: Reads approved intended-use statements, user needs, and verification results (SKILL.md). * Boundary markers: None identified. * Capability inventory: Restricted to creating and updating local documentation files (VALIDATION_PLAN.md, VALIDATION_PROTOCOL.md, VALIDATION_REPORT.md, AI_RUN_MANIFEST.yaml). * Sanitization: No explicit sanitization or filtering of external content is described.

validation-agent