builtin

Package headeredit is the built-in header_edit plugin: it sets, adds, and removes HTTP headers on the request, the client response, and the upstream provider call.

Package llmaltering is the built-in llm_based_altering plugin: it rewrites the text of selected message roles through an auxiliary model, both in the prompt phase (before the provider call) and in the response phase.

Package llmjudge is the built-in llm_judge plugin: it asks a second model whether a prompt or a completion violates a policy and blocks, answers, or flags the exchange based on the verdict.

Package presidio is the built-in presidio plugin: it sends prompt and completion text to a Presidio analyzer sidecar, and anonymizes, flags, or blocks the personal data it finds.

Package routeexample is the built-in cheapest_healthy routing strategy: it sends a virtual model's traffic to the cheapest (or fastest) target whose recent error rate is acceptable, learning target health from the attempt outcomes GoModel reports.

Package stringreplace is the built-in string_replace plugin: it rewrites, flags, or blocks prompt and completion text that matches a list of literal or regular-expression rules, in place for non-streaming responses and in flight for streams.

Package systemprompt is the built-in system_prompt plugin: it injects, overrides, or decorates the system message before the provider call.