chore(main): release LaunchDarkly.ServerSdk.Ai 0.10.0 (#278)

🤖 I have created a release *beep* *boop*
---


##
[0.10.0](LaunchDarkly.ServerSdk.Ai-v0.9.4...LaunchDarkly.ServerSdk.Ai-v0.10.0)
(2026-06-05)


### ⚠ BREAKING CHANGES

* Move Role from LaunchDarkly.Sdk.Server.Ai.DataModel into
LdAiConfigTypes
* Rename LdAiConfig.ModelConfiguration to LdAiConfigTypes.ModelConfig
and LdAiConfig.ModelProvider to LdAiConfigTypes.ProviderConfig
* Move shared types nested in LdAiConfig into static LdAiConfigTypes —
Message, ModelConfiguration, and ModelProvider
* Enforce at-most-once tracking — each metric type (duration, tokens,
feedback, success/error, time-to-first-token) records once per tracker;
duplicates are dropped with a warning
* Add ResumptionToken on ILdAiConfigTracker and
ILdAiClient.CreateTracker(resumptionToken, context) for cross-process
tracker reconstruction with the original runId
* Add per-execution runId on all AI track event payloads for billing
isolation
* Change CompletionConfig to return LdAiCompletionConfig instead of
ILdAiConfigTracker — obtain a tracker via config.CreateTracker()
* Default LdAiCompletionConfigDefault.Enabled to true per AISDK spec
(was false on the LdAiConfig builder in 0.9.x)
* Remove ILdAiConfigTracker.Config property — read config fields from
the LdAiCompletionConfig the caller already holds
* Make LdAiConfigTracker SDK-constructed only — the public constructor
is removed; obtain trackers via config.CreateTracker() or
ILdAiClient.CreateTracker(resumptionToken, context)
* Convert LdAiCompletionConfig and LdAiCompletionConfigDefault from
records to classes — equality is reference-based instead of structural
* Remove LaunchDarkly.Sdk.Server.Ai.DataModel namespace — delete unused
AiConfig, Meta, Model, Provider, and Message JSON DTO classes
* Split unified LdAiConfig into LdAiCompletionConfig (SDK output) and
LdAiCompletionConfigDefault (user input) — Builder, New(), and Disabled
move to the Default type; introduce abstract LdAiConfig and
LdAiConfigDefault base types for future agent/judge modes

### Features

* Add MetricSummary property on ILdAiConfigTracker summarizing recorded
metrics
([44ff485](44ff485))
* Add per-execution runId on all AI track event payloads for billing
isolation
([44ff485](44ff485))
* Add ResumptionToken on ILdAiConfigTracker and
ILdAiClient.CreateTracker(resumptionToken, context) for cross-process
tracker reconstruction with the original runId
([44ff485](44ff485))
* Change CompletionConfig to return LdAiCompletionConfig instead of
ILdAiConfigTracker — obtain a tracker via config.CreateTracker()
([92f799f](92f799f))
* Convert LdAiCompletionConfig and LdAiCompletionConfigDefault from
records to classes — equality is reference-based instead of structural
([92f799f](92f799f))
* Default LdAiCompletionConfigDefault.Enabled to true per AISDK spec
(was false on the LdAiConfig builder in 0.9.x)
([92f799f](92f799f))
* Enforce at-most-once tracking — each metric type (duration, tokens,
feedback, success/error, time-to-first-token) records once per tracker;
duplicates are dropped with a warning
([44ff485](44ff485))
* Make LdAiConfigTracker SDK-constructed only — the public constructor
is removed; obtain trackers via config.CreateTracker() or
ILdAiClient.CreateTracker(resumptionToken, context)
([92f799f](92f799f))
* Mode-mismatch detection — log a warning and return the caller's
default when the flag's _ldMeta.mode does not match the requested mode
(per
sdk-specs[#229](#229))
([92f799f](92f799f))
* Move Role from LaunchDarkly.Sdk.Server.Ai.DataModel into
LdAiConfigTypes
([ac7fd06](ac7fd06))
* Move shared types nested in LdAiConfig into static LdAiConfigTypes —
Message, ModelConfiguration, and ModelProvider
([ac7fd06](ac7fd06))
* Non-object variation handling — log an error and return the caller's
default when the variation result is not an object
([92f799f](92f799f))
* Per-message interpolation fallback — a malformed Mustache template
keeps raw content for that message rather than discarding the entire
config
([92f799f](92f799f))
* Remove ILdAiConfigTracker.Config property — read config fields from
the LdAiCompletionConfig the caller already holds
([92f799f](92f799f))
* Remove LaunchDarkly.Sdk.Server.Ai.DataModel namespace — delete unused
AiConfig, Meta, Model, Provider, and Message JSON DTO classes
([92f799f](92f799f))
* Rename LdAiConfig.ModelConfiguration to LdAiConfigTypes.ModelConfig
and LdAiConfig.ModelProvider to LdAiConfigTypes.ProviderConfig
([ac7fd06](ac7fd06))
* Split unified LdAiConfig into LdAiCompletionConfig (SDK output) and
LdAiCompletionConfigDefault (user input) — Builder, New(), and Disabled
move to the Default type; introduce abstract LdAiConfig and
LdAiConfigDefault base types for future agent/judge modes
([92f799f](92f799f))
* Tolerant LdValue parsing — missing or wrong-typed fields degrade to
typed defaults instead of discarding the whole config
([92f799f](92f799f))

---
This PR was generated with [Release
Please](https://github.com/googleapis/release-please). See
[documentation](https://github.com/googleapis/release-please#release-please).

<!-- CURSOR_SUMMARY -->
---

> [!NOTE]
> **High Risk**
> The tagged release ships extensive breaking public API and
telemetry/billing semantics; this PR is low-risk mechanically but
merging it publishes high consumer upgrade risk.
> 
> **Overview**
> **Release Please PR** that publishes **LaunchDarkly.ServerSdk.Ai
0.10.0** by bumping `0.9.4` → `0.10.0` in
`.release-please-manifest.json`, `LaunchDarkly.ServerSdk.Ai.csproj`,
`SdkInfo.Version`, and `PROVENANCE.md`, and adding the **0.10.0**
section to `CHANGELOG.md`.
> 
> The changelog documents a **breaking** AI SDK realignment with the
AISDK spec: split `LdAiConfig` into **`LdAiCompletionConfig`**
(evaluation output) vs **`LdAiCompletionConfigDefault`** (caller input);
**`CompletionConfig` returns config, not a tracker**
(`config.CreateTracker()` / `ILdAiClient.CreateTracker(resumptionToken,
context)`); tracker construction is SDK-only with **`runId`**,
**resumption tokens**, **at-most-once metrics**, and
**`MetricSummary`**; shared types move to **`LdAiConfigTypes`** and
**`DataModel` is removed**; completion defaults default **`Enabled` to
true**; config/records become **classes** (reference equality).
Non-breaking behavior called out includes tolerant **`LdValue`
parsing**, mode-mismatch and non-object variation fallbacks, and
per-message Mustache fallback.
> 
> <sup>Reviewed by [Cursor Bugbot](https://cursor.com/bugbot) for commit
82deb10. Bugbot is set up for automated
code reviews on this repo. Configure
[here](https://www.cursor.com/dashboard/bugbot).</sup>
<!-- /CURSOR_SUMMARY -->

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

Author: github-actions[bot]

.release-please-manifest.json

@@ -5,7 +5,7 @@
   "pkgs/sdk/server": "8.14.0",
   "pkgs/sdk/client": "5.8.0",
   "pkgs/telemetry": "1.5.1",
-  "pkgs/sdk/server-ai": "0.9.4",
+  "pkgs/sdk/server-ai": "0.10.0",
   "pkgs/shared/common": "7.2.0",
   "pkgs/shared/common-json-net": "7.0.2"
 }

pkgs/sdk/server-ai/CHANGELOG.md

@@ -1,5 +1,45 @@
 # Changelog
 
+## [0.10.0](https://github.com/launchdarkly/dotnet-core/compare/LaunchDarkly.ServerSdk.Ai-v0.9.4...LaunchDarkly.ServerSdk.Ai-v0.10.0) (2026-06-05)
+
+
+### ⚠ BREAKING CHANGES
+
+* Move Role from LaunchDarkly.Sdk.Server.Ai.DataModel into LdAiConfigTypes
+* Rename LdAiConfig.ModelConfiguration to LdAiConfigTypes.ModelConfig and LdAiConfig.ModelProvider to LdAiConfigTypes.ProviderConfig
+* Move shared types nested in LdAiConfig into static LdAiConfigTypes — Message, ModelConfiguration, and ModelProvider
+* Enforce at-most-once tracking — each metric type (duration, tokens, feedback, success/error, time-to-first-token) records once per tracker; duplicates are dropped with a warning
+* Add ResumptionToken on ILdAiConfigTracker and ILdAiClient.CreateTracker(resumptionToken, context) for cross-process tracker reconstruction with the original runId
+* Add per-execution runId on all AI track event payloads for billing isolation
+* Change CompletionConfig to return LdAiCompletionConfig instead of ILdAiConfigTracker — obtain a tracker via config.CreateTracker()
+* Default LdAiCompletionConfigDefault.Enabled to true per AISDK spec (was false on the LdAiConfig builder in 0.9.x)
+* Remove ILdAiConfigTracker.Config property — read config fields from the LdAiCompletionConfig the caller already holds
+* Make LdAiConfigTracker SDK-constructed only — the public constructor is removed; obtain trackers via config.CreateTracker() or ILdAiClient.CreateTracker(resumptionToken, context)
+* Convert LdAiCompletionConfig and LdAiCompletionConfigDefault from records to classes — equality is reference-based instead of structural
+* Remove LaunchDarkly.Sdk.Server.Ai.DataModel namespace — delete unused AiConfig, Meta, Model, Provider, and Message JSON DTO classes
+* Split unified LdAiConfig into LdAiCompletionConfig (SDK output) and LdAiCompletionConfigDefault (user input) — Builder, New(), and Disabled move to the Default type; introduce abstract LdAiConfig and LdAiConfigDefault base types for future agent/judge modes
+
+### Features
+
+* Add MetricSummary property on ILdAiConfigTracker summarizing recorded metrics ([44ff485](https://github.com/launchdarkly/dotnet-core/commit/44ff485fcc3095b3adea81d142bc865cd0904d12))
+* Add per-execution runId on all AI track event payloads for billing isolation ([44ff485](https://github.com/launchdarkly/dotnet-core/commit/44ff485fcc3095b3adea81d142bc865cd0904d12))
+* Add ResumptionToken on ILdAiConfigTracker and ILdAiClient.CreateTracker(resumptionToken, context) for cross-process tracker reconstruction with the original runId ([44ff485](https://github.com/launchdarkly/dotnet-core/commit/44ff485fcc3095b3adea81d142bc865cd0904d12))
+* Change CompletionConfig to return LdAiCompletionConfig instead of ILdAiConfigTracker — obtain a tracker via config.CreateTracker() ([92f799f](https://github.com/launchdarkly/dotnet-core/commit/92f799f6681772eb1b34dd808517ce98845f70c6))
+* Convert LdAiCompletionConfig and LdAiCompletionConfigDefault from records to classes — equality is reference-based instead of structural ([92f799f](https://github.com/launchdarkly/dotnet-core/commit/92f799f6681772eb1b34dd808517ce98845f70c6))
+* Default LdAiCompletionConfigDefault.Enabled to true per AISDK spec (was false on the LdAiConfig builder in 0.9.x) ([92f799f](https://github.com/launchdarkly/dotnet-core/commit/92f799f6681772eb1b34dd808517ce98845f70c6))
+* Enforce at-most-once tracking — each metric type (duration, tokens, feedback, success/error, time-to-first-token) records once per tracker; duplicates are dropped with a warning ([44ff485](https://github.com/launchdarkly/dotnet-core/commit/44ff485fcc3095b3adea81d142bc865cd0904d12))
+* Make LdAiConfigTracker SDK-constructed only — the public constructor is removed; obtain trackers via config.CreateTracker() or ILdAiClient.CreateTracker(resumptionToken, context) ([92f799f](https://github.com/launchdarkly/dotnet-core/commit/92f799f6681772eb1b34dd808517ce98845f70c6))
+* Mode-mismatch detection — log a warning and return the caller's default when the flag's _ldMeta.mode does not match the requested mode (per sdk-specs[#229](https://github.com/launchdarkly/dotnet-core/issues/229)) ([92f799f](https://github.com/launchdarkly/dotnet-core/commit/92f799f6681772eb1b34dd808517ce98845f70c6))
+* Move Role from LaunchDarkly.Sdk.Server.Ai.DataModel into LdAiConfigTypes ([ac7fd06](https://github.com/launchdarkly/dotnet-core/commit/ac7fd06ffa078b701f19b2eb16982103207633ff))
+* Move shared types nested in LdAiConfig into static LdAiConfigTypes — Message, ModelConfiguration, and ModelProvider ([ac7fd06](https://github.com/launchdarkly/dotnet-core/commit/ac7fd06ffa078b701f19b2eb16982103207633ff))
+* Non-object variation handling — log an error and return the caller's default when the variation result is not an object ([92f799f](https://github.com/launchdarkly/dotnet-core/commit/92f799f6681772eb1b34dd808517ce98845f70c6))
+* Per-message interpolation fallback — a malformed Mustache template keeps raw content for that message rather than discarding the entire config ([92f799f](https://github.com/launchdarkly/dotnet-core/commit/92f799f6681772eb1b34dd808517ce98845f70c6))
+* Remove ILdAiConfigTracker.Config property — read config fields from the LdAiCompletionConfig the caller already holds ([92f799f](https://github.com/launchdarkly/dotnet-core/commit/92f799f6681772eb1b34dd808517ce98845f70c6))
+* Remove LaunchDarkly.Sdk.Server.Ai.DataModel namespace — delete unused AiConfig, Meta, Model, Provider, and Message JSON DTO classes ([92f799f](https://github.com/launchdarkly/dotnet-core/commit/92f799f6681772eb1b34dd808517ce98845f70c6))
+* Rename LdAiConfig.ModelConfiguration to LdAiConfigTypes.ModelConfig and LdAiConfig.ModelProvider to LdAiConfigTypes.ProviderConfig ([ac7fd06](https://github.com/launchdarkly/dotnet-core/commit/ac7fd06ffa078b701f19b2eb16982103207633ff))
+* Split unified LdAiConfig into LdAiCompletionConfig (SDK output) and LdAiCompletionConfigDefault (user input) — Builder, New(), and Disabled move to the Default type; introduce abstract LdAiConfig and LdAiConfigDefault base types for future agent/judge modes ([92f799f](https://github.com/launchdarkly/dotnet-core/commit/92f799f6681772eb1b34dd808517ce98845f70c6))
+* Tolerant LdValue parsing — missing or wrong-typed fields degrade to typed defaults instead of discarding the whole config ([92f799f](https://github.com/launchdarkly/dotnet-core/commit/92f799f6681772eb1b34dd808517ce98845f70c6))
+
 ## [0.9.4](https://github.com/launchdarkly/dotnet-core/compare/LaunchDarkly.ServerSdk.Ai-v0.9.3...LaunchDarkly.ServerSdk.Ai-v0.9.4) (2026-05-27)
 
 

pkgs/sdk/server-ai/PROVENANCE.md

@@ -9,7 +9,7 @@ To verify build provenance attestations, we recommend using the [GitHub CLI `att
 <!-- x-release-please-start-version -->
 ```
 # Set the version of the SDK to verify
-SDK_VERSION=0.9.4
+SDK_VERSION=0.10.0
 ```
 <!-- x-release-please-end -->
 

pkgs/sdk/server-ai/src/LaunchDarkly.ServerSdk.Ai.csproj

@@ -1,7 +1,7 @@
 <Project Sdk="Microsoft.NET.Sdk">
   <PropertyGroup>
     <!--x-release-please-start-version-->
-    <Version>0.9.4</Version>
+    <Version>0.10.0</Version>
     <!--x-release-please-end-->
     <!-- The BUILDFRAMEWORKS variable allows us to override the target frameworks with a
          single framework that we are testing; this allows us to test with older SDK

pkgs/sdk/server-ai/src/SdkInfo.cs

@@ -13,7 +13,7 @@ public static class SdkInfo
     /// <summary>
     /// The version of the AI SDK package.
     /// </summary>
-    public const string Version = "0.9.4"; // x-release-please-version
+    public const string Version = "0.10.0"; // x-release-please-version
 
     /// <summary>
     /// The implementation language.