caching

Caching

Use this skill when endpoints need read-performance improvements and predictable cache behavior.

Workflow

1. Identify idempotent read endpoints (GET) suitable for caching.
2. Enable route-level caching with finite TTL.
3. Add a stable cache-key strategy when request context affects output.
4. Verify invalidation expectations against write endpoints.

Core Pattern

from litestar import get

@get("/articles", cache=60) async def list_articles() -> list[dict[str, str]]: return [{"title": "example"}]

## Strategy Notes

- Cache only safe, deterministic responses.
- Include tenant/user/locale dimensions in key building when applicable.
- Keep TTL short for frequently changing resources.
- Do not cache authenticated responses unless keying is strict and intentional.

## Litestar References

- https://docs.litestar.dev/latest/usage/caching.html
- https://docs.litestar.dev/latest/usage/responses.html#response-caching