The Agent Skills Directory

[SAFE]: The skill consists of documentation and instructional content without any executable scripts or binary files. No dangerous operations, such as arbitrary command execution or unauthorized file system access, were found.
[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to process external JSON schemas and data payloads, which is a potential surface for indirect prompt injection. However, the skill lacks any exploitable capabilities (such as tools, file-writing, or network operations) and proactively includes security guidelines to mitigate risks when handling untrusted schemas.
Ingestion points: JSON schemas and data payloads mentioned in SKILL.md instructions.
Boundary markers: None explicitly defined for prompt isolation, though the skill defines schema boundaries.
Capability inventory: None. The skill contains no code or tool definitions.
Sanitization: Instructions recommend bounding validation work and avoiding untrusted $ref targets.
[EXTERNAL_DOWNLOADS]: The skill includes reference links to well-known and trusted official documentation for JSON Schema (json-schema.org) and OpenAPI (openapis.org). These are static informational links and do not involve remote code execution.

z-json-schema