0.5.0 release with change log (#914)

0.5.0 release with change log

Author: Ololoshechkin

CHANGELOG.md

@@ -1,3 +1,121 @@
+# 0.5.0
+
+> Published 2 Oct 2025
+
+## Major Features
+
+- **Full Agent-to-Agent (A2A) Protocol Support**:
+    - **Multiplatform Kotlin A2A SDK**: Including server and client with JSON-RPC HTTP support.
+    - **A2A Agent Feature**: seamlessly integrate A2A in your Koog agents
+- **Non-Graph API for Strategies**: Introduced non-graph API for creating AI Agent strategies as Kotlin extension
+  functions with most of Koog's features supported (#560)
+- **Agent Persistence and Checkpointing**:
+    - **Roll back Tool Side-Effects**: Add `RollbackToolRegistry` in the `Persistence` feature in order to roll back
+      tool calls with side effects when checkpointing.
+    - **State-Machine Persistence / Message History Switch**: Support switching between full state-machine persistence
+      and message history persistence (#856)
+- **Tool API Improvements**:
+    - Make `ToolDescriptor` auto-generated for class-based tools (#791)
+    - Get rid of `ToolArgs` and `ToolResult` limitations for `Tool<*, *>` class (#791)
+- **`subgraphWithTask` Simplification**: Get rid of required `finishTool` and support tools as functions in
+  `subgraphWithTask`, deduce final step automatically by data class (#791)
+- **`AIAgentService` Introduced**: Make `AIAgent` state-manageable and single-run explicitly, introduce `AIAgentService`
+  to manage multiple uniform running agents.
+- **New components**:
+  - Add LLM as a Judge component (#866)
+  - Tool Calling loop with Structured Output strategy (#829)
+
+## Improvements
+
+- Make Koog-based tools exportable via MCP server (KG-388)
+- Add `additionalProperties` to LLM clients in order to support custom LLM configurations (#836)
+- Allow adjusting context window sizes for Ollama dynamically (#883)
+- Refactor streaming api to support tool calls (#747)
+- Provide an ability to collect and send a list of nodes and edges out of `AIAgentStrategy` to the client when running
+  an agent (KG-160)
+- Add `excludedProperties` to inline `createJsonStructure` too, update KDocs (#826)
+- Refactor binary attachment handling and introduce Base64 serializer (#838)
+- In `JsonStructuredData.defaultJson` instance rename class discriminator
+  from `#type` to `kind` to align with common practices (#772, KG-384)
+- Make standard json generator default when creating `JsonStructuredData`
+  (it was basic before) (#772, KG-384)
+- Add default audio configuration and modalities (#817)
+- Add `GptAudio` model in OpenAI client (#818)
+- Allow re-running of finished agents that have `Persistence` feature installed (#828, KG-193)
+- Allow ideomatic node transformations with `.transform { ...}` lambda function (#684)
+- Add ability to filter messages for every agent feature (KG-376)
+- Add support for trace-level attributes in Langfuse integration (#860, KG-427)
+- Keep all system messages when compressing message history of the agent(#857)
+- Add support for Anthropic's Sonnet 4.5 model in Anthropic/Bedrock providers (#885)
+- Refactored LLM client auto-configuration in Spring Boot integration, to modular provider-specific classes with
+  improved validation and security (#886)
+- Add LLM Streaming agent events (KG-148)
+
+## Bug Fixes
+
+- Fix broken Anthropic models support via Amazon Bedrock (#789)
+- Make `AIAgentStorageKey` in agent storage actually unique by removing `data` modifier (#825)
+- Fix rerun for agents with Persistence (#828, KG-193)
+- Update mcp version to `0.7.2` with fix for Android target (#835)
+- Do not include an empty system message in Anthropic request (#887, KG-317)
+- Use `maxTokens` from params in Google models (#734)
+- Fix finishReason nullability (#771)
+
+## Deprecations
+
+- Rename agent interceptors in `EventHandler` and related feature events (KG-376)
+- Deprecate concurrent unsafe `AIAgent.asTool` in favor of `AIAgentService.createAgentTool` (#873)
+- Rename `Persistency` to `Persistence` everywhere (#896)
+- Add `agentId` argument to all `Persistence` methods instead of `persistencyId` class field (#904)
+
+## Examples
+
+- Add a basic code-agent example (#808, KG-227)
+- Add iOS and Web targets for demo-compose-app (#779, #780)
+
+# 0.4.2
+
+> Published 15 Sep 2025
+
+## Improvements
+
+- Make agents‑mcp support KMP targets to run across more platforms (#756).
+- Add LLM client retry support to Spring Boot auto‑configuration to improve resilience on transient failures (#748).
+- Add Claude Opus 4.1 model support to Anthropic client to unlock latest reasoning capabilities (#730).
+- Add Gemini 2.5 Flash Lite model support to Google client to enable lower‑latency, cost‑efficient generations (#769).
+- Add Java‑compatible non‑streaming Prompt Executor so Java apps can call Koog without
+  coroutines ([KG-312](https://youtrack.jetbrains.com/issue/KG-312), #715).
+- Support excluding properties in JSON Schema generation to fine‑tune structured outputs (#638).
+- Update AWS SDK to latest compatible version for Bedrock integrations.
+- Introduce Postgres persistence provider to store agent state and artifacts (#705).
+- Update Kotlin to 2.2.10 in dependency configuration for improved performance and language features (#764).
+- Refactor executeStreaming to remove suspend for simpler interop and better call sites (#720).
+- Add Java‑compatible prompt executor (non‑streaming) wiring and polish across
+  modules ([KG-312](https://youtrack.jetbrains.com/issue/KG-312), #715).
+- Decouple FileSystemEntry from FileSystemProvider to simplify testing and enable alternative providers (#664).
+
+## Bug Fixes
+
+- Add missing tool calling support for Bedrock Nova models so agents can invoke functions when using
+  Nova ([KG-239](https://youtrack.jetbrains.com/issue/KG-239)).
+- Add Android target support and migrate Android app to Kotlin Multiplatform to widen KMP
+  coverage ([KG-315](https://youtrack.jetbrains.com/issue/KG-315), #728, #767).
+- Add Spring Boot Java example to jump‑start integration (#739).
+- Add Java Spring auto‑config fixes: correct property binding and make Koog starter work out of the box (#698).
+- Fix split package issues in OpenAI LLM clients to avoid classpath/load
+  errors ([KG-305](https://youtrack.jetbrains.com/issue/KG-305), #694).
+- Ensure Anthropic tool schemas include the required "type" field in serialized request bodies to prevent validation
+  errors during tool calling (#582).
+- Fix AbstractOpenAILLMClient to correctly handle plain‑text responses in capabilities flow; add integration tests to
+  prevent regressions (#564).
+- Fix GraalVM native image build failure so projects can compile native binaries again (#774).
+- Fix usages in OpenAI‑based data model to align with recent API changes (#688).
+- Fix SpringBootStarters initialization and improve `RetryingClient` (#894)
+
+## CI and Build
+
+- Nightly build configuration and dependency submission workflow added (#695, #737).
+
 # 0.4.1
 
 > Published 28 Aug 2025
@@ -13,50 +131,67 @@ Fixed iOS target publication
 ## Major Features
 
 - **Integration with Observability Tools**:
-  - **Langfuse Integration**: Span adapters for Langfuse client, including open telemetry and graph visualisation ([KG-217](https://youtrack.jetbrains.com/issue/KG-217), [KG-223](https://youtrack.jetbrains.com/issue/KG-223))
-  - **W&B Weave Integration**: Span adapters for W&B Weave open telemetry and observability ([KG-217](https://youtrack.jetbrains.com/issue/KG-217), [KG-218](https://youtrack.jetbrains.com/issue/KG-218))
-- **Ktor Integration**: First-class Ktor support via the "Koog" Ktor plugin to register and run agents in Ktor applications (#422).
-- **iOS Target Support**: Multiplatform expanded with native iOS targets, enabling agents to run on Apple platforms (#512).
-- **Upgraded Structured Output**: Refactored structured output API to be more flexible and add built-in/native provider support for OpenAI and Google, reducing prompt boilerplate and improving validation (#443).
-- **GPT5 and Custom LLM Parameters Support**: Now GPT5 is available together with custom additional LLM parameters for OpenAI-compatible clients (#631, #517)
+    - **Langfuse Integration**: Span adapters for Langfuse client, including open telemetry and graph
+      visualisation ([KG-217](https://youtrack.jetbrains.com/issue/KG-217), [KG-223](https://youtrack.jetbrains.com/issue/KG-223))
+    - **W&B Weave Integration**: Span adapters for W&B Weave open telemetry and
+      observability ([KG-217](https://youtrack.jetbrains.com/issue/KG-217), [KG-218](https://youtrack.jetbrains.com/issue/KG-218))
+- **Ktor Integration**: First-class Ktor support via the "Koog" Ktor plugin to register and run agents in Ktor
+  applications (#422).
+- **iOS Target Support**: Multiplatform expanded with native iOS targets, enabling agents to run on Apple platforms (
+  #512).
+- **Upgraded Structured Output**: Refactored structured output API to be more flexible and add built-in/native provider
+  support for OpenAI and Google, reducing prompt boilerplate and improving validation (#443).
+- **GPT5 and Custom LLM Parameters Support**: Now GPT5 is available together with custom additional LLM parameters for
+  OpenAI-compatible clients (#631, #517)
 - **Resilience and Retries**:
-  - **Retryable LLM Clients**: Introduce retry logic for LLM clients with sensible defaults to reduce transient failures (#592)
-  - **Retry Anything with LLM Feedback**: Add a feedback mechanism to the retry component (`subgraphWithRetry`) to observe and tune behavior (#459).
+    - **Retryable LLM Clients**: Introduce retry logic for LLM clients with sensible defaults to reduce transient
+      failures (#592)
+    - **Retry Anything with LLM Feedback**: Add a feedback mechanism to the retry component (`subgraphWithRetry`) to
+      observe and tune behavior (#459).
 
 ## Improvements
 
 - **OpenTelemetry and Observability**:
-  - Finish reason and unified attributes for inference/tool/message spans and events; extract event body fields to attributes for better querying ([KG-218](https://youtrack.jetbrains.com/issue/KG-218)).
-  - Mask sensitive data in events/attributes and introduce a “hidden-by-default” string type to keep secrets safe in logs ([KG-259](https://youtrack.jetbrains.com/issue/KG-259)).
-  - Include all messages into the inference span and add an index for ChoiceEvent to simplify analysis ([KG-172](https://youtrack.jetbrains.com/issue/KG-172)).
-  - Add tool arguments to `gen_ai.choice` and `gen_ai.assistant.message` events (#462).
-  - Allow setting a custom OpenTelemetry SDK instance in Koog ([KG-169](https://youtrack.jetbrains.com/issue/KG-169)).
+    - Finish reason and unified attributes for inference/tool/message spans and events; extract event body fields to
+      attributes for better querying ([KG-218](https://youtrack.jetbrains.com/issue/KG-218)).
+    - Mask sensitive data in events/attributes and introduce a “hidden-by-default” string type to keep secrets safe in
+      logs ([KG-259](https://youtrack.jetbrains.com/issue/KG-259)).
+    - Include all messages into the inference span and add an index for ChoiceEvent to simplify
+      analysis ([KG-172](https://youtrack.jetbrains.com/issue/KG-172)).
+    - Add tool arguments to `gen_ai.choice` and `gen_ai.assistant.message` events (#462).
+    - Allow setting a custom OpenTelemetry SDK instance in Koog ([KG-169](https://youtrack.jetbrains.com/issue/KG-169)).
 - **LLM and Providers**:
-  - Support Google’s “thinking” mode in generation config to improve reasoning quality (#414).
-  - Add responses API support for OpenAI (#645)
-  - AWS Bedrock: support Inference Profiles for simpler, consistent configuration (#506) and accept `AWS_SESSION_TOKEN` (#456).
-  - Add `maxTokens` as prompt parameters for finer control over generation length (#579).
-  - Add `contextLength` and `maxOutputTokens` to `LLModel` (#438, [KG-134](https://youtrack.jetbrains.com/issue/KG-134))
+    - Support Google’s “thinking” mode in generation config to improve reasoning quality (#414).
+    - Add responses API support for OpenAI (#645)
+    - AWS Bedrock: support Inference Profiles for simpler, consistent configuration (#506) and accept
+      `AWS_SESSION_TOKEN` (#456).
+    - Add `maxTokens` as prompt parameters for finer control over generation length (#579).
+    - Add `contextLength` and `maxOutputTokens` to `LLModel` (
+      #438, [KG-134](https://youtrack.jetbrains.com/issue/KG-134))
 - **Agent Engine**:
-  - Add AIAgentPipeline interceptors to uniformly handle node errors; propagate `NodeExecutionError` across features ([KG-170](https://youtrack.jetbrains.com/issue/KG-170)).
-  - Include finish node processing in the pipeline to ensure finalizers run reliably (#598).
+    - Add AIAgentPipeline interceptors to uniformly handle node errors; propagate `NodeExecutionError` across
+      features ([KG-170](https://youtrack.jetbrains.com/issue/KG-170)).
+    - Include finish node processing in the pipeline to ensure finalizers run reliably (#598).
 - **File Tools and RAG**:
-    - Reworked FileSystemProvider with API cleanups and better ergonomics; moved blocking/suspendable operations to `Dispatchers.IO` for improved performance and responsiveness (#557, “Move suspendable operations to Dispatchers.IO”).
-    - Introduce `filterByRoot` helpers and allow custom path filters in `FilteredFileSystemProvider` for safer agent sandboxes (#494, #508).
+    - Reworked FileSystemProvider with API cleanups and better ergonomics; moved blocking/suspendable operations to
+      `Dispatchers.IO` for improved performance and responsiveness (#557, “Move suspendable operations to
+      Dispatchers.IO”).
+    - Introduce `filterByRoot` helpers and allow custom path filters in `FilteredFileSystemProvider` for safer agent
+      sandboxes (#494, #508).
     - Rename `PathFilter` to `TraversalFilter` and make its methods suspendable to support async checks.
     - Rename `fromAbsoluteString` to `fromAbsolutePathString` for clarity (#567).
     - Add `ReadFileTool` for reading local file contents where appropriate (#628).
 - Update kotlin-mcp dependency to v0.6.0 (#523)
 
-
 ## Bug Fixes
 
 - Make `parts` field nullable in Google responses to handle missing content from Gemini models (#652).
 - Fix enum parsing in MCP when type is not mentioned (#601, [KG-49](https://youtrack.jetbrains.com/issue/KG-49))
 - Fix function calling for `gemini-2.5-flash` models to correctly route tool invocations (#586).
 - Restore OpenAI `responseFormat` option support in requests (#643).
 - Correct `o4-mini` vs `gpt-4o-mini` model mix-up in configuration (#573).
-- Ensure event body for function calls is valid JSON for telemetry ingestion ([KG-268](https://youtrack.jetbrains.com/issue/KG-268)).
+- Ensure event body for function calls is valid JSON for telemetry
+  ingestion ([KG-268](https://youtrack.jetbrains.com/issue/KG-268)).
 - Fix duplicated tool names resolution in `AIAgentSubgraphExt` to prevent conflicts (#493).
 - Fix Azure OpenAI client settings to generate valid endpoint URLs (#478).
 - Restore `llama3.2:latest` as the default for LLAMA_3_2 to match the provider expectations (#522).
@@ -65,14 +200,17 @@ Fixed iOS target publication
 
 ## Removals / Breaking Changes
 
-- Remove Google Gemini 1.5 Flash/Pro variants from the catalog ([KG-216](https://youtrack.jetbrains.com/issue/KG-216), #574).
+- Remove Google Gemini 1.5 Flash/Pro variants from the catalog ([KG-216](https://youtrack.jetbrains.com/issue/KG-216),
+  #574).
 - Drop `execute` extensions for `PromptExecutor` in favor of the unified API (#591).
-- File system API cleanup: removed deprecated FSProvider interfaces and methods; `PathFilter` renamed to `TraversalFilter` with suspendable operations; `fromAbsoluteString` renamed to `fromAbsolutePathString`.
+- File system API cleanup: removed deprecated FSProvider interfaces and methods; `PathFilter` renamed to
+  `TraversalFilter` with suspendable operations; `fromAbsoluteString` renamed to `fromAbsolutePathString`.
 
 ## Examples
 
 - Add a web search agent (from Koog live stream 1) showcasing retrieval + summarization (#575).
-- Add a trip planning agent example (from Koog live stream 2) demonstrating tools + planning + composite strategy (#595).
+- Add a trip planning agent example (from Koog live stream 2) demonstrating tools + planning + composite strategy (
+  #595).
 - Improve BestJokeAgent sample and fix NumberGuessingAgent example (#503, #445).
 
 # 0.3.0
@@ -86,7 +224,8 @@ Fixed iOS target publication
   the latest checkpoint (#305)
 - **Vector Document Storage**: Store embeddings and documents in persistent storage for retrieval-augmented generation (
   RAG), with in-memory and local file implementations (#272)
-- **OpenTelemetry Support**: Native integration with OpenTelemetry for unified tracing logs across AI agents (#369, #401,
+- **OpenTelemetry Support**: Native integration with OpenTelemetry for unified tracing logs across AI agents (#369,
+  #401,
   #423, #426)
 - **Content Moderation**: Built-in support for moderating models, enabling AI agents to automatically review and filter
   outputs for safety and compliance (#395)

README.md

@@ -87,7 +87,7 @@ Currently, the framework supports the JVM, JS, WasmJS and iOS targets.
 
     ```
     dependencies {
-        implementation("ai.koog:koog-agents:0.4.2")
+        implementation("ai.koog:koog-agents:0.5.0")
     }
     ```
 2. Make sure that you have `mavenCentral()` in the list of repositories.
@@ -97,7 +97,7 @@ Currently, the framework supports the JVM, JS, WasmJS and iOS targets.
 
     ```
     dependencies {
-        implementation 'ai.koog:koog-agents:0.4.2'
+        implementation 'ai.koog:koog-agents:0.5.0'
     }
     ```
 2. Make sure that you have `mavenCentral()` in the list of repositories.
@@ -109,7 +109,7 @@ Currently, the framework supports the JVM, JS, WasmJS and iOS targets.
     <dependency>
         <groupId>ai.koog</groupId>
         <artifactId>koog-agents-jvm</artifactId>
-        <version>0.4.2</version>
+        <version>0.5.0</version>
     </dependency>
     ```
 2. Make sure that you have `mavenCentral` in the list of repositories.

build.gradle.kts

@@ -18,7 +18,7 @@ group = "ai.koog"
 version = run {
     // our version follows the semver specification
 
-    val main = "0.4.3"
+    val main = "0.5.0"
 
     val feat = run {
         val releaseBuild = !System.getenv("BRANCH_KOOG_IS_RELEASING_FROM").isNullOrBlank()