agentex-sdk 0.4.11__tar.gz → 0.4.13__tar.gz

agentex_sdk-0.4.13/.release-please-manifest.json

@@ -0,0 +1,3 @@
+{
+  ".": "0.4.13"
+}

{agentex_sdk-0.4.11 → agentex_sdk-0.4.13}/CHANGELOG.md

@@ -1,5 +1,26 @@
 # Changelog
 
+## 0.4.13 (2025-09-12)
+
+Full Changelog: [v0.4.12...v0.4.13](https://github.com/scaleapi/agentex-python/compare/v0.4.12...v0.4.13)
+
+### Features
+
+* **api:** api update ([0102183](https://github.com/scaleapi/agentex-python/commit/0102183a8f5a23dbdaf905ffbe7ffbcf59bf7b21))
+* **api:** api update ([8a6edb1](https://github.com/scaleapi/agentex-python/commit/8a6edb13046ca24bf6c45fc018e32de498d48869))
+
+## 0.4.12 (2025-09-08)
+
+Full Changelog: [v0.4.11...v0.4.12](https://github.com/scaleapi/agentex-python/compare/v0.4.11...v0.4.12)
+
+### ⚠ BREAKING CHANGES
+
+* task_cancel now requires explicit agent_name/agent_id parameter to identify which agent owns the task being cancelled
+
+### Bug Fixes
+
+* task cancellation architectural bug ([f9a72a9](https://github.com/scaleapi/agentex-python/commit/f9a72a94f96afe86d3cc80f4f85ea368279d4517))
+
 ## 0.4.11 (2025-09-04)
 
 Full Changelog: [v0.4.10...v0.4.11](https://github.com/scaleapi/agentex-python/compare/v0.4.10...v0.4.11)

agentex_sdk-0.4.13/CLAUDE.md

@@ -0,0 +1,93 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## Development Commands
+
+### Package Management in the top level repo
+- Use `rye` for dependency management (preferred)
+- Run `./scripts/bootstrap` to set up the environment
+- Or use `rye sync --all-features` directly
+
+Special note: the individual tutorials maintain their own tutorial specific virtualenv using `uv`. So when testing/running tutorials, you `uv run` instead of `rye run`.  Everything else is similar.
+
+#### Testing
+- Run tests: `rye run pytest` or `./scripts/test`
+- Run specific test: `rye run pytest path/to/test_file.py::TestClass::test_method -v`
+- Mock server is automatically started for tests, runs on port 4010
+
+#### Linting and Formatting
+- Format code: `rye run format` or `./scripts/format`
+  * The repository is still in flux, so running format might accidentally change files that aren't part of your scope of changes. So always run `run rye format` with additional arguments to constrain the formatting to the files that you are modifying.
+- Lint code: `rye run lint` or `./scripts/lint`
+- Type check: `rye run typecheck` (runs both pyright and mypy)
+
+### Building and Running
+- Build package: `rye build`
+
+
+
+### CLI Commands
+The package provides the `agentex` CLI with these main commands:
+- `agentex agents` - Get, list, run, build, and deploy agents
+- `agentex tasks` - Get, list, and delete tasks  
+- `agentex secrets` - Sync, get, list, and delete secrets
+- `agentex uv` - UV wrapper with AgentEx-specific enhancements
+- `agentex init` - Initialize new agent projects
+
+### Agent Development
+- Run agents: `agentex agents run --manifest manifest.yaml`
+- Debug agents: `agentex agents run --manifest manifest.yaml --debug-worker`
+- Debug with custom port: `agentex agents run --manifest manifest.yaml --debug-worker --debug-port 5679`
+
+## Architecture Overview
+
+### Code Structure
+- `/src/agentex/` - Core SDK and generated API client code
+- `/src/agentex/lib/` - Custom library code (not modified by code generator)
+  - `/cli/` - Command-line interface implementation
+  - `/core/` - Core services, adapters, and temporal workflows
+  - `/sdk/` - SDK utilities and FastACP implementation
+  - `/types/` - Custom type definitions
+  - `/utils/` - Utility functions
+- `/examples/` - Example implementations and tutorials
+- `/tests/` - Test suites
+
+### Key Components
+
+**SDK Architecture:**
+- **Client Layer**: HTTP client for AgentEx API (`_client.py`, `resources/`)
+- **CLI Layer**: Typer-based command interface (`lib/cli/`)
+- **Core Services**: Temporal workflows, adapters, and services (`lib/core/`)
+- **FastACP**: Fast Agent Communication Protocol implementation (`lib/sdk/fastacp/`)
+- **State Machine**: Workflow state management (`lib/sdk/state_machine/`)
+
+**Temporal Integration:**
+- Workflow definitions in `lib/core/temporal/`
+- Activity definitions for different providers
+- Worker implementations for running temporal workflows
+
+**Agent Framework:**
+- Manifest-driven agent configuration
+- Support for multiple agent types (sync, temporal-based)
+- Debugging support with VS Code integration
+
+### Code Generation
+Most SDK code is auto-generated. Manual changes are preserved in:
+- `src/agentex/lib/` directory
+- `examples/` directory
+- Merge conflicts may occur between manual patches and generator changes
+
+### Key Dependencies
+- `temporalio` - Temporal workflow engine
+- `typer` - CLI framework  
+- `pydantic` - Data validation
+- `httpx` - HTTP client
+- `fastapi` - Web framework
+- `ruff` - Linting and formatting
+- `pytest` - Testing framework
+
+### Environment Requirements
+- Python 3.12+ required
+- Uses Rye for dependency management
+- Supports both sync and async client patterns

{agentex_sdk-0.4.11 → agentex_sdk-0.4.13}/CONTRIBUTING.md

@@ -126,3 +126,33 @@ You can release to package managers by using [the `Publish PyPI` GitHub action](
 
 If you need to manually release a package, you can run the `bin/publish-pypi` script with a `PYPI_TOKEN` set on
 the environment.
+
+## 🤖 **Vibe Coding Setup**
+
+This repository is setup with some pre-canned prompts for [Claude Code](https://docs.anthropic.com/en/docs/claude-code) as well as [Cursor](https://cursor.com/).
+
+### Cursor
+
+Access to Cursor can be acquired by asking for it in #it-help.  Then just loading this repo in the Cursor IDE should enable the prompts.
+
+### Claude Code
+
+### 1. Install Claude Code
+```bash
+npm install -g @anthropic-ai/claude-code
+```
+
+### 2. Request a LiteLLM API Key
+Visit the [LiteLLM User Guide](https://scale.atlassian.net/wiki/spaces/EPD/pages/1490354189/LiteLLM+User+Guide#Requesting-LiteLLM-Key-for-Generic-Usage) to request your API key.
+
+### 3. Set Environment Variables
+```bash
+export ANTHROPIC_AUTH_TOKEN=${LITELLM_PROXY_API_KEY}
+export ANTHROPIC_BASE_URL="https://litellm.ml-serving-internal.scale.com"
+```
+
+### 4. Start Claude Code
+```bash
+claude
+```
+This should be run from inside the main repo directory. If you run the command from a terminal inside VSCode, then Claude will use the VSCode editor to show diffs etc. 

{agentex_sdk-0.4.11 → agentex_sdk-0.4.13}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: agentex-sdk
-Version: 0.4.11
+Version: 0.4.13
 Summary: The official Python library for the agentex API
 Project-URL: Homepage, https://github.com/scaleapi/agentex-python
 Project-URL: Repository, https://github.com/scaleapi/agentex-python

{agentex_sdk-0.4.11 → agentex_sdk-0.4.13}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "agentex-sdk"
-version = "0.4.11"
+version = "0.4.13"
 description = "The official Python library for the agentex API"
 dynamic = ["readme"]
 license = "Apache-2.0"

{agentex_sdk-0.4.11 → agentex_sdk-0.4.13}/src/agentex/_constants.py

@@ -6,9 +6,9 @@ RAW_RESPONSE_HEADER = "X-Stainless-Raw-Response"
 OVERRIDE_CAST_TO_HEADER = "____stainless_override_cast_to"
 
 # default timeout is 1 minute
-DEFAULT_TIMEOUT = httpx.Timeout(timeout=60, connect=5.0)
-DEFAULT_MAX_RETRIES = 2
-DEFAULT_CONNECTION_LIMITS = httpx.Limits(max_connections=100, max_keepalive_connections=20)
+DEFAULT_TIMEOUT = httpx.Timeout(timeout=300, connect=5.0)
+DEFAULT_MAX_RETRIES = 0
+DEFAULT_CONNECTION_LIMITS = httpx.Limits(max_connections=1000, max_keepalive_connections=1000)
 
 INITIAL_RETRY_DELAY = 0.5
 MAX_RETRY_DELAY = 8.0

{agentex_sdk-0.4.11 → agentex_sdk-0.4.13}/src/agentex/_version.py

@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "agentex"
-__version__ = "0.4.11"  # x-release-please-version
+__version__ = "0.4.13"  # x-release-please-version

{agentex_sdk-0.4.11 → agentex_sdk-0.4.13}/src/agentex/lib/adk/_modules/acp.py

@@ -59,6 +59,7 @@ class ACPModule:
         start_to_close_timeout: timedelta = timedelta(seconds=5),
         heartbeat_timeout: timedelta = timedelta(seconds=5),
         retry_policy: RetryPolicy = DEFAULT_RETRY_POLICY,
+        request: dict[str, Any] | None = None,
     ) -> Task:
         """
         Create a new task.
@@ -71,6 +72,7 @@ class ACPModule:
             start_to_close_timeout: The start to close timeout for the task.
             heartbeat_timeout: The heartbeat timeout for the task.
             retry_policy: The retry policy for the task.
+            request: Additional request context including headers to forward to the agent.
 
         Returns:
             The task entry.
@@ -85,6 +87,7 @@ class ACPModule:
                     params=params,
                     trace_id=trace_id,
                     parent_span_id=parent_span_id,
+                    request=request,
                 ),
                 response_type=Task,
                 start_to_close_timeout=start_to_close_timeout,
@@ -99,6 +102,7 @@ class ACPModule:
                 params=params,
                 trace_id=trace_id,
                 parent_span_id=parent_span_id,
+                request=request,
             )
 
     async def send_event(
@@ -112,15 +116,22 @@ class ACPModule:
         start_to_close_timeout: timedelta = timedelta(seconds=5),
         heartbeat_timeout: timedelta = timedelta(seconds=5),
         retry_policy: RetryPolicy = DEFAULT_RETRY_POLICY,
+        request: dict[str, Any] | None = None,
     ) -> Event:
         """
         Send an event to a task.
 
         Args:
             task_id: The ID of the task to send the event to.
-            data: The data to send to the event.
+            content: The content to send to the event.
             agent_id: The ID of the agent to send the event to.
             agent_name: The name of the agent to send the event to.
+            trace_id: The trace ID for the event.
+            parent_span_id: The parent span ID for the event.
+            start_to_close_timeout: The start to close timeout for the event.
+            heartbeat_timeout: The heartbeat timeout for the event.
+            retry_policy: The retry policy for the event.
+            request: Additional request context including headers to forward to the agent.
 
         Returns:
             The event entry.
@@ -135,6 +146,7 @@ class ACPModule:
                     content=content,
                     trace_id=trace_id,
                     parent_span_id=parent_span_id,
+                    request=request,
                 ),
                 response_type=None,
                 start_to_close_timeout=start_to_close_timeout,
@@ -149,6 +161,7 @@ class ACPModule:
                 content=content,
                 trace_id=trace_id,
                 parent_span_id=parent_span_id,
+                request=request,
             )
 
     async def send_message(
@@ -162,15 +175,22 @@ class ACPModule:
         start_to_close_timeout: timedelta = timedelta(seconds=5),
         heartbeat_timeout: timedelta = timedelta(seconds=5),
         retry_policy: RetryPolicy = DEFAULT_RETRY_POLICY,
+        request: dict[str, Any] | None = None,
     ) -> List[TaskMessage]:
         """
         Send a message to a task.
 
         Args:
-            task_id: The ID of the task to send the message to.
             content: The task message content to send to the task.
+            task_id: The ID of the task to send the message to.
             agent_id: The ID of the agent to send the message to.
             agent_name: The name of the agent to send the message to.
+            trace_id: The trace ID for the message.
+            parent_span_id: The parent span ID for the message.
+            start_to_close_timeout: The start to close timeout for the message.
+            heartbeat_timeout: The heartbeat timeout for the message.
+            retry_policy: The retry policy for the message.
+            request: Additional request context including headers to forward to the agent.
 
         Returns:
             The message entry.
@@ -185,6 +205,7 @@ class ACPModule:
                     content=content,
                     trace_id=trace_id,
                     parent_span_id=parent_span_id,
+                    request=request,
                 ),
                 response_type=TaskMessage,
                 start_to_close_timeout=start_to_close_timeout,
@@ -199,32 +220,43 @@ class ACPModule:
                 content=content,
                 trace_id=trace_id,
                 parent_span_id=parent_span_id,
+                request=request,
             )
 
     async def cancel_task(
         self,
         task_id: str | None = None,
         task_name: str | None = None,
+        agent_id: str | None = None,
+        agent_name: str | None = None,
         trace_id: str | None = None,
         parent_span_id: str | None = None,
         start_to_close_timeout: timedelta = timedelta(seconds=5),
         heartbeat_timeout: timedelta = timedelta(seconds=5),
         retry_policy: RetryPolicy = DEFAULT_RETRY_POLICY,
+        request: dict[str, Any] | None = None,
     ) -> Task:
         """
-        Cancel a task.
+        Cancel a task by sending cancel request to the agent that owns the task.
 
         Args:
-            task_id: The ID of the task to cancel.
-            task_name: The name of the task to cancel.
+            task_id: ID of the task to cancel.
+            task_name: Name of the task to cancel.
+            agent_id: ID of the agent that owns the task.
+            agent_name: Name of the agent that owns the task.
             trace_id: The trace ID for the task.
             parent_span_id: The parent span ID for the task.
             start_to_close_timeout: The start to close timeout for the task.
             heartbeat_timeout: The heartbeat timeout for the task.
             retry_policy: The retry policy for the task.
+            request: Additional request context including headers to forward to the agent.
 
         Returns:
             The task entry.
+            
+        Raises:
+            ValueError: If neither agent_name nor agent_id is provided,
+                       or if neither task_name nor task_id is provided
         """
         if in_temporal_workflow():
             return await ActivityHelpers.execute_activity(
@@ -232,8 +264,11 @@ class ACPModule:
                 request=TaskCancelParams(
                     task_id=task_id,
                     task_name=task_name,
+                    agent_id=agent_id,
+                    agent_name=agent_name,
                     trace_id=trace_id,
                     parent_span_id=parent_span_id,
+                    request=request,
                 ),
                 response_type=None,
                 start_to_close_timeout=start_to_close_timeout,
@@ -244,6 +279,9 @@ class ACPModule:
             return await self._acp_service.task_cancel(
                 task_id=task_id,
                 task_name=task_name,
+                agent_id=agent_id,
+                agent_name=agent_name,
                 trace_id=trace_id,
                 parent_span_id=parent_span_id,
+                request=request,
             )

{agentex_sdk-0.4.11 → agentex_sdk-0.4.13}/src/agentex/lib/adk/providers/_modules/openai.py

@@ -88,6 +88,7 @@ class OpenAIModule:
         mcp_timeout_seconds: int | None = None,
         input_guardrails: list[InputGuardrail] | None = None,
         output_guardrails: list[OutputGuardrail] | None = None,
+        max_turns: int | None = None,
     ) -> SerializableRunResult | RunResult:
         """
         Run an agent without streaming or TaskMessage creation.
@@ -114,6 +115,7 @@ class OpenAIModule:
             mcp_timeout_seconds: Optional param to set the timeout threshold for the MCP servers. Defaults to 5 seconds.
             input_guardrails: Optional list of input guardrails to run on initial user input.
             output_guardrails: Optional list of output guardrails to run on final agent output.
+            max_turns: Maximum number of turns the agent can take. Uses Runner's default if None.
 
         Returns:
             Union[SerializableRunResult, RunResult]: SerializableRunResult when in Temporal, RunResult otherwise.
@@ -136,6 +138,7 @@ class OpenAIModule:
                 mcp_timeout_seconds=mcp_timeout_seconds,
                 input_guardrails=input_guardrails,
                 output_guardrails=output_guardrails,
+                max_turns=max_turns,
             )
             return await ActivityHelpers.execute_activity(
                 activity_name=OpenAIActivityName.RUN_AGENT,
@@ -163,6 +166,7 @@ class OpenAIModule:
                 mcp_timeout_seconds=mcp_timeout_seconds,
                 input_guardrails=input_guardrails,
                 output_guardrails=output_guardrails,
+                max_turns=max_turns,
             )
 
     async def run_agent_auto_send(
@@ -191,6 +195,7 @@ class OpenAIModule:
         mcp_timeout_seconds: int | None = None,
         input_guardrails: list[InputGuardrail] | None = None,
         output_guardrails: list[OutputGuardrail] | None = None,
+        max_turns: int | None = None,
     ) -> SerializableRunResult | RunResult:
         """
         Run an agent with automatic TaskMessage creation.
@@ -216,6 +221,7 @@ class OpenAIModule:
             mcp_timeout_seconds: Optional param to set the timeout threshold for the MCP servers. Defaults to 5 seconds.
             input_guardrails: Optional list of input guardrails to run on initial user input.
             output_guardrails: Optional list of output guardrails to run on final agent output.
+            max_turns: Maximum number of turns the agent can take. Uses Runner's default if None.
 
         Returns:
             Union[SerializableRunResult, RunResult]: SerializableRunResult when in Temporal, RunResult otherwise.
@@ -239,6 +245,7 @@ class OpenAIModule:
                 mcp_timeout_seconds=mcp_timeout_seconds,
                 input_guardrails=input_guardrails,
                 output_guardrails=output_guardrails,
+                max_turns=max_turns,
             )
             return await ActivityHelpers.execute_activity(
                 activity_name=OpenAIActivityName.RUN_AGENT_AUTO_SEND,
@@ -267,6 +274,7 @@ class OpenAIModule:
                 mcp_timeout_seconds=mcp_timeout_seconds,
                 input_guardrails=input_guardrails,
                 output_guardrails=output_guardrails,
+                max_turns=max_turns,
             )
 
     async def run_agent_streamed(
@@ -291,6 +299,7 @@ class OpenAIModule:
         mcp_timeout_seconds: int | None = None,
         input_guardrails: list[InputGuardrail] | None = None,
         output_guardrails: list[OutputGuardrail] | None = None,
+        max_turns: int | None = None,
     ) -> RunResultStreaming:
         """
         Run an agent with streaming enabled but no TaskMessage creation.
@@ -320,6 +329,7 @@ class OpenAIModule:
             mcp_timeout_seconds: Optional param to set the timeout threshold for the MCP servers. Defaults to 5 seconds.
             input_guardrails: Optional list of input guardrails to run on initial user input.
             output_guardrails: Optional list of output guardrails to run on final agent output.
+            max_turns: Maximum number of turns the agent can take. Uses Runner's default if None.
 
         Returns:
             RunResultStreaming: The result of the agent run with streaming.
@@ -352,6 +362,7 @@ class OpenAIModule:
             mcp_timeout_seconds=mcp_timeout_seconds,
             input_guardrails=input_guardrails,
             output_guardrails=output_guardrails,
+            max_turns=max_turns,
         )
 
     async def run_agent_streamed_auto_send(
@@ -380,6 +391,7 @@ class OpenAIModule:
         mcp_timeout_seconds: int | None = None,
         input_guardrails: list[InputGuardrail] | None = None,
         output_guardrails: list[OutputGuardrail] | None = None,
+        max_turns: int | None = None,
     ) -> SerializableRunResultStreaming | RunResultStreaming:
         """
         Run an agent with streaming enabled and automatic TaskMessage creation.
@@ -405,6 +417,7 @@ class OpenAIModule:
             output_type: Optional output type.
             tool_use_behavior: Optional tool use behavior.
             mcp_timeout_seconds: Optional param to set the timeout threshold for the MCP servers. Defaults to 5 seconds.
+            max_turns: Maximum number of turns the agent can take. Uses Runner's default if None.
 
         Returns:
             Union[SerializableRunResultStreaming, RunResultStreaming]: SerializableRunResultStreaming when in Temporal, RunResultStreaming otherwise.
@@ -428,6 +441,7 @@ class OpenAIModule:
                 mcp_timeout_seconds=mcp_timeout_seconds,
                 input_guardrails=input_guardrails,
                 output_guardrails=output_guardrails,
+                max_turns=max_turns
             )
             return await ActivityHelpers.execute_activity(
                 activity_name=OpenAIActivityName.RUN_AGENT_STREAMED_AUTO_SEND,
@@ -456,4 +470,5 @@ class OpenAIModule:
                 mcp_timeout_seconds=mcp_timeout_seconds,
                 input_guardrails=input_guardrails,
                 output_guardrails=output_guardrails,
+                max_turns=max_turns,
             )

{agentex_sdk-0.4.11 → agentex_sdk-0.4.13}/src/agentex/lib/cli/handlers/deploy_handlers.py

@@ -229,9 +229,12 @@ def merge_deployment_configs(
                 all_env_vars[EnvVarKeys.AUTH_PRINCIPAL_B64.value] = encoded_principal
             else:
                 raise DeploymentError(f"Auth principal unable to be encoded for agent_env_config: {agent_env_config}")
-                
+        
+        logger.info(f"Defined agent helm overrides: {agent_env_config.helm_overrides}")
+        logger.info(f"Before-merge helm values: {helm_values}")
         if agent_env_config.helm_overrides:
             _deep_merge(helm_values, agent_env_config.helm_overrides)
+        logger.info(f"After-merge helm values: {helm_values}")
 
     # Set final environment variables
     # Environment variable precedence: manifest -> environments.yaml -> secrets (highest)

{agentex_sdk-0.4.11 → agentex_sdk-0.4.13}/src/agentex/lib/cli/templates/temporal/environments.yaml.j2

@@ -52,7 +52,7 @@ environments:
         limits:
           cpu: "1000m"
           memory: "2Gi"
-      temporal:
+      temporal-worker:
         enabled: true
         replicaCount: 2
         resources: