IO Mapper Agent

badge

About the IO Mapper Agent

When connecting agents in an application, the output of an agent needs to be compatible with the input of the agent that is connected to it. This compatibility needs to be guaranteed at three different levels:

  1. Transport level: the two agents need to use the same transport protocol.

  2. Format level: the two agents need to carry information using the same format (for example, the same JSON data structures).

  3. Semantic level: the two agents need to “talk about the same thing”.

Communication between agents is not possible if there are discrepancies between the agents at any of these levels.

Ensuring that agents are semantically compatible, that is, the output of the one agent contains the information needed by later agents, is an problem of composition or planning in the application. The IO Mapper Agent addresses level 2 and 3 compatibility. It is a component, implemented as an agent, that can make use of an LLM to transform the output of one agent to become compatible to the input of another agent. This can mean many different things:

  • JSON structure transcoding: A JSON dictionary needs to be remapped into another JSON dictionary.

  • Text summarisation: A text needs to be summarised or some information needs to be removed.

  • Text translation: A text needs to be translated from one language to another.

  • Text manipulation: Part of the information of one text needs to be reformulated into another text.

  • A combination of the above.

The IO mapper Agent can be fed the schema definitions of inputs and outputs as defined by the Agent Connect Protocol.

Getting Started

Use in your project

pip install agntcy-iomapper

To get a local copy up and running, follow the steps below.

Prerequisites

  1. Clone the repository

git clone https://github.com/agntcy/iomapper-agnt.git

  1. Install dependecies

poetry install

Usage

There are several different ways to leverage the IO Mapper functions in Python. There is an How to use the Agent IO mapping using models that can be invoked on different AI platforms and an imperative interface that does deterministic JSON remapping without using any AI models.

Key Features

The IO Mapper Agent uses an LLM to transform the inputs (typically the output of an agent) to match the desired output (typically the input of another agent). As such, it additionally supports specifying the model prompts for the translation. The configuration object provides a specification for the system and default user prompts:

This project supports specifying model interactions using LangGraph.

How to use the Agent IO mapping

Note

For each example, the detailed process of creating agents and configuring the respective multi-agent software is omitted. Instead, only the essential steps for configuring and integrating the IO mapper agent are presented.

LangGraph

We support usages with both LangGraph state defined with TypedDict or as a Pydantic object

Entities

class IOMappingAgentMetadata
Description:

The IOMappingAgentMetadata class is a data model that extends BaseModel. It is designed to facilitate the mapping process by defining input and output fields, schemas, and mapping configurations. Below is the detailed documentation of each attribute within the class.

Attributes:

This class has the following properties

input_fields (List[Union[str, FieldMetadata]]) [Required]

An array of JSON paths representing fields to be used by the IO mapper in the mapping.

Example

input_fields = ["state.fiedl1", "state.field2", FieldMetadata(json_path="state", description="this is a list of items")]

output_fields (List[Union[str, FieldMetadata]]) [Required]

An array of JSON paths representing fields to be used by the IO mapper in the result.

Example

output_fields = ["state.output_fiedl1"]

input_schema (Optional[dict[str, Any]]) [Optional]

Defines the schema for the input data.

Example

input_schema = {
    "type": "object",
    "properties": {
        "title": {"type": "string"},
        "ingredients": {"type": "array", "items": {"type": "string"}},
        "instructions": {"type": "string"},
    },
    "required": ["title", "ingredients, instructions"],
}

output_schema (Optional[dict[str, Any]]) [Optional]

Defines the schema for the result of the mapping.

Example

output_schema = {"name": {"type": "string"}, "age": {"type": "integer"}}

field_mapping (Optional[dict[str, Union[str, Callable]]]) [Optional]

A dictionary representing how the imperative mapping should be done, where the keys are fields of the output object and values are JSONPath strings or callables.

Example

field_mapping = {
    "output_name": "$.data.name",
    "output_age": lambda x: x["data"]["age"] + 1,
}
class IOMappingAgent
Description:

This class exposes all functionalities needed for IO mapping in multi-agent systems.

Attributes:

This class has the following properties

metadata (Optional[IOMappingAgentMetadata])[Required] Represents, details about the fields to be used in the translation and about the output.

Example

Here is an example of a value for metadata:

metadata_example = IOMappingAgentMetadata(
  input_fields=["documents.0.page_content"],
  output_fields=["recipe"],
  input_schema=TypeAdapter(GraphState).json_schema(),
  output_schema={
      "type": "object",
      "properties": {
          "title": {"type": "string"},
          "ingredients": {"type": "array", "items": {"type": "string"}},
          "instructions": {"type": "string"},
      },
      "required": ["title", "ingredients, instructions"],
  },
)

llm (Optional[Union[BaseChatModel, str]]) [Optional]

Model to use for translation as LangChain description or model class.

Example

Here is an example of a value for llm:

llm_example = AzureChatOpenAI(
      model=model_version,
      api_version=api_version,
      seed=42,
      temperature=0,
)

LangGraph Example 1

This example involves a multi-agent software system designed to process a create engagement campaign and share within an organization. It interacts with an agent specialized in creating campaigns, another agent specialized in identifying suitable users. The information is then relayed to an IO mapper, which converts the list of users and the campaign details to present statistics about the campaign.

Define an agent io mapper metadata

metadata = IOMappingAgentMetadata(
    input_fields=["selected_users", "campaign_details.name"],
    output_fields=["stats.status"],
)

The above instruction directs the IO mapper agent to utilize the selected_users and name from the campaign_details field and map them to the stats.status. No further information is needed since the type information can be derived from the input data which is a pydantic model.

Tip

Both input_fields and output_fields can also be sourced with a list composed of str and/or instances of FieldMetadata as the bellow example shows

metadata = IOMappingAgentMetadata(
    input_fields=[
        FieldMetadata(
            json_path="selected_users", description="A list of users to be targeted"
        ),
        FieldMetadata(
            json_path="campaign_details.name",
            description="The name that can be used by the campaign",
            examples=["Campaign A"]
        ),
    ],
    output_fields=["stats"],
)

Define an Instance of the Agent

mapping_agent = IOMappingAgent(metadata=metadata, llm=llm)

Add the node to the LangGraph graph

Add the Edge

With the edge added, you can run the your LangGraph graph.

workflow.add_edge("create_communication", "io_mapping")
workflow.add_edge("io_mapping", "send_communication")

LangGraph Example 2

This example involves a multi-agent software system designed to process a list of ingredients. It interacts with an agent specialized in recipe books to identify feasible recipes based on the provided ingredients. The information is then relayed to an IO mapper, which converts it into a format suitable for display to the user.

Define an Agent IO Mapper Metadata

metadata = IOMappingAgentMetadata(
    input_fields=["documents.0.page_content"],
    output_fields=["recipe"],
    input_schema=TypeAdapter(GraphState).json_schema(),
    output_schema={
        "type": "object",
        "properties": {
            "title": {"type": "string"},
            "ingredients": {"type": "array", "items": {"type": "string"}},
            "instructions": {"type": "string"},
        },
        "required": ["title", "ingredients, instructions"],
    },
)

Define an Instance of the Agent

mapping_agent = IOMappingAgent(metadata=metadata, llm=llm)

Add the node to the LangGraph graph

graph.add_node(
    "recipe_io_mapper",
    mapping_agent.langgraph_node,
)

Add the Edge

With the edge added, you can run the your LangGraph graph.

graph.add_edge("recipe_expert", "recipe_io_mapper")

LlamaIndex

We support both LlamaIndex Workflow and the new AgentWorkflow multi agent software

Entities

class IOMappingInputEvent
Description:

Handles input events related to IO mapping.

Attributes:

This class has the following attributes

metadata (IOMappingAgentMetadata) [Required]

This object represents information relative to input fields, output fields, and other IO mapping-related information.

Example

Here is an example of a value for metadata:

metadata_example = IOMappingAgentMetadata(
    input_fields=["field1", "field2"],
    output_fields=["field3", "field4"]
)

config (LLamaIndexIOMapperConfig) [Required] This object contains information such as the LLM instance that will be used to perform the translation.

Example

Here is an example of a value for config:

config_example = LLamaIndexIOMapperConfig(llm=llm)

data (:py:class: Any) [Required] Represents the input data to be used in the translation.

Example

Here is an example of a value for data:

data_example = {
    "text": "Translate this text into French."
}
class IOMappingOutputEvent
Description:

Handles output events related to IO mapping.

Attributes:

This class has the following attributes

mapping_result (dict)[Required] This is where the mapping result will be populated.

Example of usage in a LlamaIndex workflow

In this example we recreate the campaign workflow using LlamaIndex workflow

Begin by importing the neccessary object

from agntcy_iomapper import IOMappingAgent, IOMappingAgentMetadata

Define the workflow

 1class CampaignWorkflow(Workflow):
 2    @step
 3    async def prompt_step(self, ctx: Context, ev: StartEvent) -> PickUsersEvent:
 4        await ctx.set("llm", ev.get("llm"))
 5        return PickUsersEvent(prompt=ev.get("prompt"))
 6
 7    @step
 8    async def pick_users_step(
 9        self, ctx: Context, ev: PickUsersEvent
10    ) -> CreateCampaignEvent:
11        return CreateCampaignEvent(list_users=users)
12
13    # The step that will trigger IO mapping
14    @step
15    async def create_campaign(
16        self, ctx: Context, ev: CreateCampaignEvent
17    ) -> IOMappingInputEvent:
18        prompt = f"""
19        You are a campaign builder for company XYZ. Given a list of selected users and a user prompt, create an engaging campaign.
20        Return the campaign details as a JSON object with the following structure:
21        {{
22            "name": "Campaign Name",
23            "content": "Campaign Content",
24            "is_urgent": yes/no
25        }}
26        Selected Users: {ev.list_users}
27        User Prompt: Create a campaign for all users
28        """
29        parser = PydanticOutputParser(output_cls=Campaign)
30        llm = await ctx.get("llm", default=None)
31
32        llm_response = llm.complete(prompt)
33        try:
34            campaign_details = parser.parse(str(llm_response))
35            metadata = IOMappingAgentMetadata(
36                input_fields=["selected_users", "campaign_details.name"],
37                output_fields=["stats"],
38            )
39            config = LLamaIndexIOMapperConfig(llm=llm)
40
41            io_mapping_input_event = IOMappingInputEvent(
42                metadata=metadata,
43                config=config,
44                data=OverallState(
45                    campaign_details=campaign_details,
46                    selected_users=ev.list_users,
47                ),
48            )
49            return io_mapping_input_event
50        except Exception as e:
51            print(f"Error parsing campaign details: {e}")
52            return StopEvent(result=f"{e}")
53
54    @step
55    async def after_translation(self, evt: IOMappingOutputEvent) -> StopEvent:
56        return StopEvent(result="Done")

Tip

The highlighted lines shows how the io mapper can be triggered

Add The IO mapper step

w = CampaignWorkflow()
IOMappingAgent.as_worfklow_step(workflow=w)

Example of usage in a LlamaIndex AgentWorkflow

In this example we recreate the recipe workflow using LlamaIndex AgentWorkflow

Import the necessary objects

from agntcy_iomapper import FieldMetadata, IOMappingAgent, IOMappingAgentMetadata

Define an instance of the IOMappingAgentMetadata

mapping_metadata = IOMappingAgentMetadata(
    input_fields=["documents.0.text"],
    output_fields=[
        FieldMetadata(
            json_path="recipe",
            description="this is a recipe for the ingredients you've provided",
        )
    ],
    input_schema=TypeAdapter(GraphState).json_schema(),
    output_schema={
        "type": "object",
        "properties": {
            "title": {"type": "string"},
            "ingredients": {"type": "array", "items": {"type": "string"}},
            "instructions": {"type": "string"},
        },
        "required": ["title", "ingredients, instructions"],
    },
)

Finally define the IOMappingAgent and add it to the AgentWorkflow.

Important to note that a tool is passed, to instruct the io mapper where to go next in the flow.

io_mapping_agent = IOMappingAgent.as_workflow_agent(
    mapping_metadata=mapping_metadata,
    llm=llm,
    name="IOMapperAgent",
    description="Useful for mapping a recipe document into recipe object",
    can_handoff_to=["Formatter_Agent"],
    tools=[got_to_format],
)


io_mapping_agent = IOMappingAgent.as_workflow_agent(
    mapping_metadata=mapping_metadata,
    llm=llm,
    name="IOMapperAgent",
    description="Useful for mapping a recipe document into recipe object",
    can_handoff_to=["Formatter_Agent"],
    tools=[got_to_format],
)

Use Examples

  1. Install:

  1. From the examples folder run the desired make command, for example:

make make run_lg_eg_py

Contributing

Contributions are what make the open source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated. For detailed contributing guidelines, please see CONTRIBUTING.md