docs: refresh google cached content refs

docs/concepts/model-providers.md

@@ -257,6 +257,9 @@ OpenClaw ships with the pi‑ai catalog. These providers require **no**
 - Example models: `google/gemini-3.1-pro-preview`, `google/gemini-3-flash-preview`
 - Compatibility: legacy OpenClaw config using `google/gemini-3.1-flash-preview` is normalized to `google/gemini-3-flash-preview`
 - CLI: `openclaw onboard --auth-choice gemini-api-key`
+- Direct Gemini runs also accept `agents.defaults.models["google/<model>"].params.cachedContent`
+  (or legacy `cached_content`) to forward a provider-native
+  `cachedContents/...` handle; Gemini cache hits surface as OpenClaw `cacheRead`
 
 ### Google Vertex and Gemini CLI
 

docs/providers/google.md

@@ -83,6 +83,36 @@ retry.
 | Web search (Grounding) | Yes               |
 | Thinking/reasoning     | Yes (Gemini 3.1+) |
 
+## Direct Gemini cache reuse
+
+For direct Gemini API runs (`api: "google-generative-ai"`), OpenClaw now
+passes a configured `cachedContent` handle through to Gemini requests.
+
+- Configure per-model or global params with either
+  `cachedContent` or legacy `cached_content`
+- If both are present, `cachedContent` wins
+- Example value: `cachedContents/prebuilt-context`
+- Gemini cache-hit usage is normalized into OpenClaw `cacheRead` from
+  upstream `cachedContentTokenCount`
+
+Example:
+
+```json5
+{
+  agents: {
+    defaults: {
+      models: {
+        "google/gemini-2.5-pro": {
+          params: {
+            cachedContent: "cachedContents/prebuilt-context",
+          },
+        },
+      },
+    },
+  },
+}
+```
+
 ## Image generation
 
 The bundled `google` image-generation provider defaults to

docs/reference/prompt-caching.md

@@ -127,6 +127,17 @@ stops injecting those OpenRouter-specific Anthropic cache markers.
 
 If the provider does not support this cache mode, `cacheRetention` has no effect.
 
+### Google Gemini direct API
+
+- Direct Gemini transport (`api: "google-generative-ai"`) reports cache hits
+  through upstream `cachedContentTokenCount`; OpenClaw maps that to `cacheRead`.
+- If you already have a Gemini cached-content handle, you can pass it through as
+  `params.cachedContent` (or legacy `params.cached_content`) on the configured
+  model.
+- This is separate from Anthropic/OpenAI prompt-prefix caching. OpenClaw is
+  forwarding a provider-native cached-content reference, not synthesizing cache
+  markers.
+
 ## OpenClaw cache-stability guards
 
 OpenClaw also keeps several cache-sensitive payload shapes deterministic before