The Agent Skills Directory

[NO_CODE]: The skill consists entirely of instructional markdown and reference files without any executable scripts or binary files.
[SAFE]: External references are limited to official vendor domains (api.legnext.ai and docs.legnext.ai) for API interactions and schema validation.
[SAFE]: The integration methodology promotes security best practices, including the use of environment variables for API keys and the protection of job IDs as sensitive capability tokens.
[PROMPT_INJECTION]: The skill facilitates the processing of user-supplied prompt strings for image generation, representing an indirect prompt injection surface.
Ingestion points: User-provided text submitted to the POST /api/v1/diffusion endpoint, as documented in SKILL.md.
Boundary markers: The production checklist in references/production-checklist.md explicitly advises treating user-supplied prompts as untrusted data.
Capability inventory: The skill instructs the agent to perform network requests to the Legnext API, as described in SKILL.md and references/task-lifecycle.md.
Sanitization: references/production-checklist.md provides guidance to treat prompts as untrusted data, specifically cautioning against the use of eval() on user inputs and warning of SSRF risks from external image URLs.

legnext