This is an automated email from the ASF dual-hosted git repository.
xintongsong pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/flink-agents.git
The following commit(s) were added to refs/heads/main by this push:
new a26532e1 [Docs] Cross-Language Actions (#714)
a26532e1 is described below
commit a26532e1a1fd560bfdc4eecb11a5587c44b418a7
Author: Adesh Nalpet Adimurthy <[email protected]>
AuthorDate: Sun May 31 03:01:48 2026 -0400
[Docs] Cross-Language Actions (#714)
---
docs/content/docs/development/workflow_agent.md | 52 +++++++++++++++++++++++++
docs/content/docs/development/yaml.md | 2 +-
2 files changed, 53 insertions(+), 1 deletion(-)
diff --git a/docs/content/docs/development/workflow_agent.md
b/docs/content/docs/development/workflow_agent.md
index fa0f4af9..01073d27 100644
--- a/docs/content/docs/development/workflow_agent.md
+++ b/docs/content/docs/development/workflow_agent.md
@@ -498,6 +498,58 @@ To use async execution on JDK 21+, user should append jvm
option `--add-exports=
{{< /tab >}}
{{< /tabs >}}
+### Cross-language Actions
+
+An action declared in one language can dispatch its body to the other language
by setting a `target` on the decorator/annotation. The decorated function or
annotated method then acts as a stub — it should raise so direct calls outside
the framework fail loud.
+
+{{< tabs "Cross-language Actions" >}}
+
+{{< tab "Python" >}}
+```python
+from flink_agents.api.function import JavaFunction
+
+class MyAgent(Agent):
+ @action(
+ InputEvent.EVENT_TYPE,
+ target=JavaFunction(
+ qualname="com.example.MyHandlers",
+ method_name="handleInput",
+ parameter_types=[
+ "org.apache.flink.agents.api.Event",
+ "org.apache.flink.agents.api.context.RunnerContext",
+ ],
+ ),
+ )
+ @staticmethod
+ def handle_input(event: Event, ctx: RunnerContext) -> None:
+ raise NotImplementedError("cross-language stub")
+```
+{{< /tab >}}
+
+{{< tab "Java" >}}
+```java
+public class MyAgent extends Agent {
+ @Action(
+ listenEventTypes = {InputEvent.EVENT_TYPE},
+ target = @PythonFunction(
+ module = "my_pkg.handlers",
+ qualname = "handle_input"))
+ public static void handleInput(Event event, RunnerContext ctx) {
+ throw new UnsupportedOperationException("cross-language stub");
+ }
+}
+```
+{{< /tab >}}
+
+{{< /tabs >}}
+
+{{< hint warning >}}
+**Limitations:**
+
+- Cross-language actions are currently supported only when [running in
Flink]({{< ref "docs/operations/deployment#run-in-flink" >}}), not in local
development mode
+- Complex object serialization between languages may have limitations
+{{< /hint >}}
+
## Event
Events are JSON-serializable messages passed between actions. Every event has
a `type` string used for routing and an `attributes` map that carries the
payload. A single event may trigger multiple actions if they are all listening
to its type.
diff --git a/docs/content/docs/development/yaml.md
b/docs/content/docs/development/yaml.md
index 9bea5c46..30760d80 100644
--- a/docs/content/docs/development/yaml.md
+++ b/docs/content/docs/development/yaml.md
@@ -525,7 +525,7 @@ Loaded with `agents_env.load_yaml(...)` on the Python side,
this produces an age
- `math_chat_model` is a Python Ollama setup that calls a Java function tool
through the cross-language tool bridge.
- `creative_chat_model` is a Java Ollama setup driven from the Python loader
via the Java chat-model wrapper.
-Not every resource type is cross-language. Currently `chat_model_connections`,
`chat_model_setups`, `embedding_model_connections`, `embedding_model_setups`,
and `vector_stores` support `type:` on the opposite language; others (e.g.
`mcp_servers`) do not.
+Not every resource type is cross-language. Currently `actions`, `tools`,
`chat_model_connections`, `chat_model_setups`, `embedding_model_connections`,
`embedding_model_setups`, and `vector_stores` support `type:` on the opposite
language; others (e.g. `mcp_servers`) do not.
## YAML API Specification
