The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes user-supplied regulatory phrasing to perform impact analysis, which constitutes an indirect prompt injection surface. * Ingestion points: User-supplied text for regulation changes in Step 1 of the workflow. * Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the workflow instructions. * Capability inventory: Read and write access to local case-specific markdown files; no network or arbitrary shell execution capabilities were detected. * Sanitization: Input text is used directly for diffing and search without explicit sanitization or escaping mechanisms.
[SAFE]: The overall behavior is consistent with the stated purpose of regulatory document management. * No unauthorized network connections or external data transmissions were detected. * File system interactions are restricted to the agent's specific configuration directories and active project workspaces. * The skill does not attempt to execute remote scripts, perform privilege escalation, or establish persistence.

handbook-updates