The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides mechanisms to process user-controlled input from HTMX-specific HTTP headers.
Ingestion points: HX-Prompt and Triggering-Event headers are exposed through request.htmx in SKILL.md.
Boundary markers: The instructions emphasize separating full-page routes from fragment routes but do not specify input delimiters.
Capability inventory: The skill facilitates template responses and response header manipulation across litestar_htmx.request and litestar_htmx.response.
Sanitization: Includes a specific guardrail mandating template usage to ensure automatic HTML escaping and explicitly forbids returning unsanitized, concatenated HTML.
[SAFE]: The skill includes security best practices for CSRF protection and emphasizes the importance of using application-level security features when processing state-changing HTMX requests.
[SAFE]: All external URLs and dependencies trace back to official litestar-org repositories, PyPI, or the official HTMX project documentation.

litestar-htmx