The Agent Skills Directory

[SAFE]: The skill acts as a documentation and schema reference for the JSON Canvas format. It does not include executable code, script commands, or requests for elevated system permissions.
[EXTERNAL_DOWNLOADS]: The skill contains references to the official JSON Canvas specification and its public GitHub repository. These are well-known and trusted resources for this file format and do not constitute a security risk.
[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for processing external .canvas files which may contain untrusted user content in text nodes (ingestion point: SKILL.md). While the skill does not define specific boundary markers or sanitization logic, it describes a data-formatting capability rather than instruction execution, posing no elevated risk beyond standard data processing.

json-canvas