Update kagent 1.x reference docs at v1.0.0-alpha8 (version conrefs changed)

docs-site/assets/kagent-docs/versions/agent-substrate.md

@@ -1 +1 @@
-{{< version include-if="0.x" >}}0.0.9{{< /version >}}{{< version include-if="1.x" >}}0.3.0-alpha3{{< /version >}}
+{{< version include-if="0.x" >}}0.0.9{{< /version >}}{{< version include-if="1.x" >}}0.4.0-alpha1{{< /version >}}

docs-site/assets/kagent-docs/versions/kagent.md

@@ -1 +1 @@
-{{< version include-if="0.x" >}}0.10.2{{< /version >}}{{< version include-if="1.x" >}}1.0.0-alpha7{{< /version >}}
+{{< version include-if="0.x" >}}0.10.2{{< /version >}}{{< version include-if="1.x" >}}1.0.0-alpha8{{< /version >}}

docs-site/assets/kagent-docs/versions/runtime-image-claude.md

@@ -1 +1 @@
-ghcr.io/kagent-dev/kagent/claude-harness@sha256:1bce31de2ffe651c23f1fc64d57d6625d937ad4e8e46836f4f2208128f2c2bb0
+ghcr.io/kagent-dev/kagent/claude-harness@sha256:5f7071cb7934ca769902fb9ba94af175eb6a2fa6163e82b69b686e6532ec04b5

docs-site/assets/kagent-docs/versions/runtime-image.md

@@ -1 +1 @@
-ghcr.io/kagent-dev/kagent/golang-adk@sha256:215417b5401310bb496ae1687bb8622f93fd19991a218c6a218987836e71da84
+ghcr.io/kagent-dev/kagent/golang-adk@sha256:10ef1350786b5d21bc002ed54833fd0272ea7cd8182a92977686b7bafa7f67d3

docs-site/content/kagent/1.x/reference/api-ref.md

@@ -164,6 +164,8 @@ _Appears in:_
 | `temperature` _string_ | Temperature for sampling |  |  |
 | `topP` _string_ | Top-p sampling parameter |  |  |
 | `topK` _integer_ | Top-k sampling parameter |  |  |
+| `promptCaching` _boolean_ | PromptCaching enables Anthropic prompt caching by marking the reusable<br />prefix of every Messages API request with `cache_control` breakpoints:<br />the last tool definition, the last system prompt block and the last<br />content block of the most recent conversation turn. Anthropic caches the<br />prefix up to each breakpoint and bills a cache hit at a fraction of the<br />normal input price on later requests within the TTL. Because the<br />conversation breakpoint moves with every turn, each call of an agent<br />loop reads the whole previous history from the cache and writes only<br />the new turn.<br />Recommended for tool-using agents that make many model calls per task<br />with a stable system prompt and tool set — without it the full history<br />is billed as fresh input on every call. Cache writes are billed at a<br />premium over normal input, so a prefix has to be read at least once to<br />pay off, and each model has a minimum cacheable prefix (1024–4096<br />tokens depending on the model) below which the markers are silently<br />ignored.<br />See https://platform.claude.com/docs/en/build-with-claude/prompt-caching<br />for the current list of supported models, minimum prefix sizes and<br />pricing. | false |  |
+| `cacheTTL` _string_ | CacheTTL controls how long Anthropic retains a cached prefix when<br />PromptCaching is enabled. Only meaningful when PromptCaching is true.<br />  - "5m" (default): the standard 5-minute cache. Each cache hit refreshes<br />    the window, so an agent loop whose calls are less than 5 minutes<br />    apart keeps its prefix cached for the whole task.<br />  - "1h": the extended 1-hour cache, useful for tasks whose model calls<br />    are spaced more than 5 minutes apart.<br />NOTE: "1h" is NOT strictly better than "5m". 1-hour cache writes are<br />billed at a higher per-token rate than 5-minute writes. Only choose<br />"1h" when calls are spaced far enough apart that a 5-minute cache would<br />expire between them; otherwise the higher write cost is wasted. See the<br />Anthropic prompt-caching docs above. | 5m | Enum: [5m 1h] <br /> |
 
 #### AnthropicVertexAIConfig
 

docs-site/content/kagent/1.x/reference/helm.md

@@ -12,7 +12,7 @@ A Helm chart for kagent, built with Google ADK
 | Repository | Name | Version |
 |------------|------|---------|
 | `${SUBSTRATE_REPO}` | substrate | `${SUBSTRATE_VERSION}` |
-| file://../tools/grafana-mcp | grafana-mcp | 1.0.0-alpha7 |
+| file://../tools/grafana-mcp | grafana-mcp | 1.0.0-alpha8 |
 | https://oauth2-proxy.github.io/manifests | oauth2-proxy | ~10.7.0 |
 | oci://ghcr.io/kagent-dev/kmcp/helm | kmcp | `${KMCP_VERSION}` |
 | oci://ghcr.io/kagent-dev/tools/helm | kagent-tools | 0.3.0 |
@@ -84,6 +84,7 @@ A Helm chart for kagent, built with Google ADK
 | controller.serviceAccount.create | bool | `true` | Create the controller ServiceAccount. Set to false to use one managed outside the chart. |
 | controller.serviceAccount.name | string | `<fullname>-controller` | Name of the controller ServiceAccount. |
 | controller.sessionIdleTTL | string | `"168h"` |  |
+| controller.sessionShareMaxTTL | string | `"0"` |  |
 | controller.startupProbe | object | httpGet /health on port http, periodSeconds=15, initialDelaySeconds=15 | Custom startup probe for the controller container. Setting a value replaces the default probe entirely — include a handler (httpGet / exec / tcpSocket / grpc) when overriding. |
 | controller.streaming | string | `nil` | @deprecated Removed in 0.10.0. The A2A SDK now handles SSE buffering and timeouts internally. These values have no effect and will be removed in a future release. |
 | controller.substrate.ateApiEndpoint | string | `""` |  |