The Agent Skills Directory

[SAFE]: The skill's behavior is consistent with its stated purpose of managing website copy. No malicious intent, obfuscation, or safety bypasses were found.- [DATA_EXFILTRATION]: The skill operates entirely on local project files (e.g., BRAND.md, BRIEF.md) and does not utilize any network-capable tools or access sensitive credentials.- [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection due to its ingestion of external project files and route strings.
Ingestion points: Reads BRAND.md, BRIEF.md, and route strings in SKILL.md.
Boundary markers: None identified.
Capability inventory: Capability to write to page-spec and project code files.
Sanitization: No explicit sanitization or filtering is defined for the ingested content.

copy-pass