Design coded DeepAgents harness

samples/README.md

@@ -6,6 +6,9 @@ This sample demonstrates a simple LangGraph agent that performs basic arithmetic
 ## [Chat agent](chat-agent)
 This sample shows how to build an AI assistant using LangGraph and Tavily search for movie research and recommendations.
 
+## [Coded DeepAgent](coded-deepagent)
+This sample demonstrates a task-mode coded DeepAgent using the UiPath runtime workspace contract, typed input/output, a tool, and a review subagent.
+
 ## [Company research agent](company-research-agent)
 This sample demonstrates how to create an agent that researches companies and develops outreach strategies using web search capabilities.
 

samples/coded-deepagent/README.md

@@ -0,0 +1,43 @@
+# Coded DeepAgent
+
+This sample demonstrates a task-mode coded agent built with the UiPath
+DeepAgents contract.
+
+The graph uses `create_uipath_deep_agent_graph`, which tags the graph for the
+UiPath LangGraph runtime. At runtime, UiPath creates a workspace, injects its
+path into LangGraph config, and hydrates that workspace through job
+attachments. Task-mode DeepAgents persist workspace changes when the run
+successfully completes or suspends. The sample does not configure a custom
+bucket backend.
+
+## What It Shows
+
+- Typed coded-agent input and output with Pydantic models.
+- System and user prompts rendered from typed input.
+- A standard LangChain tool used by the main DeepAgent.
+- A DeepAgents subagent used for risk review.
+- Runtime-provided workspace persistence through the UiPath contract.
+
+## Files
+
+- `graph.py`: task-mode coded DeepAgent graph.
+- `input.json`: sample input payload.
+- `langgraph.json`: LangGraph entrypoint.
+- `uipath.json`: UiPath task-mode runtime configuration.
+- `agent.mermaid`: high-level graph diagram.
+
+## Requirements
+
+- UiPath runtime credentials for `UiPathChat`.
+- Access to the configured model, `gpt-4o-2024-08-06`.
+
+## Run
+
+```bash
+cd samples/coded-deepagent
+uv sync
+uipath run agent "$(cat input.json)"
+```
+
+The agent writes `/launch/brief.md` and `/launch/risks.md` in the DeepAgents
+workspace and returns those paths in `workspace_files`.

samples/coded-deepagent/agent.mermaid

@@ -0,0 +1,15 @@
+flowchart TB
+  __start__(__start__)
+  transform_input(transform_input)
+  advanced_agent(advanced_agent)
+  risk_reviewer([risk_reviewer subagent])
+  workspace[(runtime workspace)]
+  transform_output(transform_output)
+  __end__(__end__)
+
+  __start__ --> transform_input
+  transform_input --> advanced_agent
+  advanced_agent -. delegates review .-> risk_reviewer
+  advanced_agent -. writes files .-> workspace
+  advanced_agent --> transform_output
+  transform_output --> __end__

samples/coded-deepagent/graph.py

@@ -0,0 +1,98 @@
+"""Task-mode coded DeepAgent using the UiPath DeepAgents contract."""
+
+from langchain_core.tools import tool
+from pydantic import BaseModel, Field
+
+from uipath_langchain.chat import UiPathChat
+from uipath_langchain.deepagents import create_uipath_deep_agent_graph
+
+
+class BriefInput(BaseModel):
+    """Input for the product launch brief."""
+
+    product_name: str = Field(description="Name of the product or feature.")
+    audience: str = Field(description="Primary customer or user audience.")
+    objective: str = Field(description="Main launch objective.")
+    constraints: list[str] = Field(
+        default_factory=list,
+        description="Important limits, requirements, or risks to account for.",
+    )
+
+
+class BriefOutput(BaseModel):
+    """Structured launch brief returned by the agent."""
+
+    executive_summary: str
+    launch_plan: list[str]
+    risk_review: list[str]
+    workspace_files: list[str]
+
+
+@tool
+def score_launch_readiness(audience: str, constraints: list[str]) -> str:
+    """Score launch readiness from simple deterministic planning signals."""
+    score = 80
+    if len(constraints) >= 3:
+        score -= 10
+    if any("compliance" in item.lower() for item in constraints):
+        score -= 10
+    if any("deadline" in item.lower() for item in constraints):
+        score -= 5
+    return f"Launch readiness for {audience}: {max(score, 35)}/100"
+
+
+MODEL = UiPathChat(model="gpt-4o-2024-08-06", temperature=0)
+
+RISK_REVIEWER_PROMPT = """You are a launch risk reviewer.
+Review the proposed launch plan for practical delivery risks, compliance gaps,
+unclear ownership, and missing follow-up work. Return concise findings that the
+main agent can incorporate into the final brief."""
+
+RISK_REVIEWER = {
+    "name": "risk_reviewer",
+    "description": "Reviews launch plans for execution risks and missing safeguards.",
+    "system_prompt": RISK_REVIEWER_PROMPT,
+    "model": MODEL,
+}
+
+
+def build_system_prompt(args: dict) -> str:
+    return f"""You are a product launch planning agent for {args["product_name"]}.
+
+Use the planning tools provided by DeepAgents. Use score_launch_readiness to get
+a deterministic readiness signal. Delegate a plan review to risk_reviewer before
+finalizing the answer.
+
+Tailor all planning to this audience: {args["audience"]}.
+
+The UiPath runtime provides your DeepAgents filesystem through the tagged graph
+contract. Write these workspace files before producing the final answer:
+- /launch/brief.md with the final launch brief
+- /launch/risks.md with the risk review
+
+Return structured output matching the schema. Include the workspace file paths
+you wrote in workspace_files."""
+
+
+def build_user_prompt(args: dict) -> str:
+    constraints = args.get("constraints") or []
+    constraint_lines = "\n".join(f"- {item}" for item in constraints) or "- None"
+    return f"""Create a launch brief.
+
+Product: {args["product_name"]}
+Audience: {args["audience"]}
+Objective: {args["objective"]}
+Constraints:
+{constraint_lines}
+"""
+
+
+graph = create_uipath_deep_agent_graph(
+    model=MODEL,
+    input_schema=BriefInput,
+    output_schema=BriefOutput,
+    system_prompt=build_system_prompt,
+    user_prompt=build_user_prompt,
+    tools=[score_launch_readiness],
+    subagents=[RISK_REVIEWER],
+)

samples/coded-deepagent/input.json

@@ -0,0 +1,10 @@
+{
+  "product_name": "Invoice Exception Copilot",
+  "audience": "accounts payable operations managers",
+  "objective": "prepare a pilot launch plan for three enterprise customers",
+  "constraints": [
+    "security review must complete before external rollout",
+    "deadline is six weeks away",
+    "support team needs enablement material"
+  ]
+}

samples/coded-deepagent/langgraph.json

@@ -0,0 +1,7 @@
+{
+  "dependencies": ["."],
+  "graphs": {
+    "agent": "./graph.py:graph"
+  },
+  "env": ".env"
+}

samples/coded-deepagent/pyproject.toml

@@ -0,0 +1,15 @@
+[project]
+name = "coded-deepagent"
+version = "0.0.1"
+description = "Task-mode coded DeepAgent using the UiPath runtime workspace contract"
+authors = [{ name = "UiPath", email = "support@uipath.com" }]
+requires-python = ">=3.11"
+dependencies = [
+    "uipath-langchain>=0.14.3, <0.15.0",
+    "uipath>=2.13.6, <2.14.0",
+]
+
+[dependency-groups]
+dev = [
+    "uipath-dev",
+]

samples/coded-deepagent/uipath.json

@@ -0,0 +1,5 @@
+{
+  "runtimeOptions": {
+    "isConversational": false
+  }
+}

src/uipath_langchain/agent/advanced/agent.py

@@ -6,11 +6,11 @@
 from deepagents import CompiledSubAgent, SubAgent
 from deepagents import create_deep_agent as _create_deep_agent
 from deepagents.backends import BackendProtocol
-from deepagents.backends.filesystem import FilesystemBackend
 from deepagents.backends.protocol import BackendFactory
 from langchain.agents.structured_output import ResponseFormat
 from langchain_core.language_models import BaseChatModel
-from langchain_core.messages import HumanMessage
+from langchain_core.messages import HumanMessage, SystemMessage
+from langchain_core.runnables.config import RunnableConfig
 from langchain_core.tools import BaseTool
 from langgraph.graph import END, START
 from langgraph.graph.state import CompiledStateGraph, StateGraph
@@ -23,6 +23,7 @@
 from .utils import (
     MEMORY_INDEX_VIRTUAL_PATH,
     create_state_with_input,
+    is_workspace_filesystem_backend,
     resolve_input_attachments,
 )
 
@@ -62,22 +63,25 @@ def create_advanced_agent_graph(
     input_schema: type[BaseModel] | None,
     output_schema: type[BaseModel],
     build_user_message: Callable[[dict[str, Any]], str],
+    build_system_message: Callable[[dict[str, Any]], str] | None = None,
+    subagents: Sequence[SubAgent | CompiledSubAgent] = (),
 ) -> StateGraph[Any, Any, Any, Any]:
     """Wrap the advanced agent in a parent graph that maps typed I/O to/from messages.
 
-    With a ``FilesystemBackend``, attachment-shaped inputs are downloaded into the
-    workspace and given a ``FilePath`` before the user message is built. A
-    ``FilesystemBackend`` also enables workspace memory: deepagents'
+    With a filesystem workspace backend, attachment-shaped inputs are downloaded
+    into the workspace and given a ``FilePath`` before the user message is built.
+    Filesystem workspaces also enable workspace memory: deepagents'
     ``MemoryMiddleware`` reads ``/memory/MEMORY.md`` from the backend each turn.
-    Memory stays disabled for non-filesystem backends, which carry no workspace.
+    Memory stays disabled for backends which carry no workspace.
     """
     memory_sources = (
-        [MEMORY_INDEX_VIRTUAL_PATH] if isinstance(backend, FilesystemBackend) else []
+        [MEMORY_INDEX_VIRTUAL_PATH] if is_workspace_filesystem_backend(backend) else []
     )
 
     inner_graph = create_advanced_agent(
         model=model,
         tools=tools,
+        subagents=subagents,
         system_prompt=system_prompt,
         backend=backend,
         response_format=response_format,
@@ -90,7 +94,10 @@ def create_advanced_agent_graph(
         get_job_attachment_paths(input_schema) if input_schema is not None else []
     )
 
-    async def transform_input_async(state: BaseModel) -> dict[str, Any]:
+    async def transform_input_async(
+        state: BaseModel,
+        config: RunnableConfig,
+    ) -> dict[str, Any]:
         state_data = state.model_dump()
         input_data = {k: v for k, v in state_data.items() if k not in internal_fields}
         input_args = (
@@ -100,10 +107,20 @@ async def transform_input_async(state: BaseModel) -> dict[str, Any]:
         )
         if attachment_paths:
             input_args = await resolve_input_attachments(
-                backend, attachment_paths, input_args
+                backend,
+                attachment_paths,
+                input_args,
+                state=state,
+                config=config,
             )
+        messages = []
+        if build_system_message is not None:
+            system_text = build_system_message(input_args)
+            if system_text:
+                messages.append(SystemMessage(content=system_text, id="system-input"))
         user_text = build_user_message(input_args)
-        return {"messages": [HumanMessage(content=user_text, id="user-input")]}
+        messages.append(HumanMessage(content=user_text, id="user-input"))
+        return {"messages": messages}
 
     def transform_output(state: BaseModel) -> dict[str, Any]:
         structured = getattr(state, "structured_response", {})
@@ -128,6 +145,7 @@ def create_conversational_advanced_agent_graph(
     tools: Sequence[BaseTool],
     system_prompt: str,
     backend: BackendProtocol | BackendFactory | None,
+    subagents: Sequence[SubAgent | CompiledSubAgent] = (),
 ) -> StateGraph[Any, Any, Any, Any]:
     """Wrap the advanced agent in a parent graph that speaks the conversational contract.
 
@@ -141,12 +159,13 @@ def create_conversational_advanced_agent_graph(
     from uipath_langchain.runtime.messages import UiPathChatMessagesMapper
 
     memory_sources = (
-        [MEMORY_INDEX_VIRTUAL_PATH] if isinstance(backend, FilesystemBackend) else []
+        [MEMORY_INDEX_VIRTUAL_PATH] if is_workspace_filesystem_backend(backend) else []
     )
 
     inner_graph = create_advanced_agent(
         model=model,
         tools=tools,
+        subagents=subagents,
         system_prompt=system_prompt,
         backend=backend,
         memory=memory_sources,