​
Basic Geometric Example

A simple example showing how to use objectHashes.
agent.py
from typing import Annotated

from encord.objects.ontology_labels_impl import LabelRowV2
from encord.objects.ontology_object_instance import ObjectInstance
from fastapi import Depends, FastAPI

from encord_agents.fastapi.cors import get_encord_app()
from encord_agents.fastapi.dependencies import (
    FrameData,
    dep_label_row,
    dep_objects,
)

# Initialize FastAPI app
app = get_encord_app()


@app.post("/handle-object-hashes")
def handle_object_hashes(
    frame_data: FrameData,
    lr: Annotated[LabelRowV2, Depends(dep_label_row)],
    object_instances: Annotated[list[ObjectInstance], Depends(dep_objects)],
) -> None:
    for object_inst in object_instances:
        print(object_inst)
Use Case: Selective OCR on Selected Objects This functionality allows you to apply your own OCR model to specific objects selected directly within the Encord platform. When you trigger your agent from the Encord app after selecting objects, the platform automatically sends a list of objectHashes to your agent. Your agent can then use the dep_objects method to gain immediate access to these specific object instances, which greatly simplifies integrating your OCR model for targeted processing. Test the Agent
  1. Save the above code as agent.py.
  2. Run the following command to run the agent in debug mode in your terminal.
uvicorn main:app --reload --port 8080
  1. Open your Project in the Encord platform and navigate to a frame with an object that you want to act on. Choose an object from the bottom left sider and click Copy URL as shown:
Copy URL from left sider
The url should have roughly this format: "https://app.encord.com/label_editor/{project_hash}/{data_hash}/{frame}/0?other_query_params&objectHash={objectHash}".
  1. In another shell operating from the same working directory, source your virtual environment and test the agent.
source venv/bin/activate
encord-agents test local agent '<your_url>'
  1. To see if the test is successful, refresh your browser to see the action taken by the Agent. If the test has run successfully, the agent can be deployed. Visit the deployment documentation to learn more.

​
Nested Classification using Claude 3.5 Sonnet

The goals of this example is to:
  1. Create an editor agent that can automatically fill in frame-level classifications in the Label Editor.
  2. Demonstrate how to use the OntologyDataModel for classifications.
  3. Demonstrate how to build an agent using FastAPI that can be self-hosted.
Prerequisites Before you begin, ensure you have: Run the following commands to set up your environment: 
python -m venv venv                   # Create a virtual Python environment  
source venv/bin/activate              # Activate the virtual environment 
python -m pip install "fastapi[standard]" encord-agents anthropic # Install required dependencies  
export ANTHROPIC_API_KEY="<your_api_key>" # Set your Anthropic API key 
export ENCORD_SSH_KEY_FILE="/path/to/your/private/key"  # Define your Encord access key
Project Setup Create a Project with visual content (images, image groups, image sequences, or videos) in Encord. This example uses the following Ontology, but any Ontology containing classifications can be used.
Ontology JSON
{
  "objects": [],
  "classifications": [
    {
      "id": "1",
      "featureNodeHash": "TTkHMtuD",
      "attributes": [
        {
          "id": "1.1",
          "featureNodeHash": "+1g9I9Sg",
          "type": "text",
          "name": "scene summary",
          "required": false,
          "dynamic": false
        }
      ]
    },
    {
      "id": "2",
      "featureNodeHash": "xGV/wCD0",
      "attributes": [
        {
          "id": "2.1",
          "featureNodeHash": "k3EVexk7",
          "type": "radio",
          "name": "is there a person in the frame?",
          "required": false,
          "options": [
            {
              "id": "2.1.1",
              "featureNodeHash": "EkGwhcO4",
              "label": "yes",
              "value": "yes",
              "options": [
                {
                  "id": "2.1.1.1",
                  "featureNodeHash": "mj9QCDY4",
                  "type": "text",
                  "name": "What is the person doing?",
                  "required": false
                }
              ]
            },
            {
              "id": "2.1.2",
              "featureNodeHash": "37rMLC/v",
              "label": "no",
              "value": "no",
              "options": []
            }
          ],
          "dynamic": false
        }
      ]
    }
  ]
}
To construct the same Ontology as used in this example, run the following script.
Create Ontology
import json
from encord.objects.ontology_structure import OntologyStructure
from encord_agents.core.utils import get_user_client

encord_client = get_user_client()
structure = OntologyStructure.from_dict(json.loads("{the_json_above}"))
ontology = encord_client.create_ontology(
    title="Your ontology title",
    structure=structure
)
print(ontology.ontology_hash)
The aim is to trigger an agent that transforms a labeling task from Figure A to Figure B. Figure A: No classification labels. Figure B: Multiple nested classification labels generated by an LLM. Create the Agent This section provides the complete code for creating your editor agent, along with an explanation of its internal workings. Agent Setup Steps
  1. Import dependencies, authenticate with Encord, and set up the Project. Ensure you insert your Project’s unique identifier.
  2. Create a data model and a system prompt based on the Project Ontology to tell Claude how to structure its response.
  3. Set up an Anthropic API client to establish communication with the Claude model.
  4. Define the Editor Agent. This includes:
  • Receiving frame data using FastAPI’s Form dependency.
  • Retrieving the associated label row and frame content using Encord Agents’ dependencies.
  • Constructing a Frame object from the content.
  • Sending the frame image to Claude for analysis.
  • Parsing Claude’s response into classification instances.
  • Adding these classifications to the label row and saving the updated data.
# 1. Import dependencies and set up the Project. The CORS middleware is crucial as it allows the Encord platform to make requests to your API.
import os

import numpy as np
from anthropic import Anthropic
from encord.objects.ontology_labels_impl import LabelRowV2
from fastapi import Depends
from numpy.typing import NDArray
from typing_extensions import Annotated

from encord_agents.core.data_model import Frame
from encord_agents.core.ontology import OntologyDataModel
from encord_agents.core.utils import get_user_client
from encord_agents.fastapi.cors import get_encord_app
from encord_agents.fastapi.dependencies import (
    FrameData,
    dep_label_row,
    dep_single_frame,
)

# Initialize FastAPI app
app = get_encord_app()

# 2. Set up the Project and create a data model based on the Ontology.

client = get_user_client()
project = client.get_project("<your_project_hash>")
data_model = OntologyDataModel(project.ontology_structure.classifications)

# 3. Set up Claude and create the system prompt that tells Claude how to structure its response.
system_prompt = f"""
You're a helpful assistant that's supposed to help fill in json objects 
according to this schema:

    ```json
    {data_model.model_json_schema_str}
    ```

Please only respond with valid json.
"""

ANTHROPIC_API_KEY = os.getenv("ANTHROPIC_API_KEY")
anthropic_client = Anthropic(api_key=ANTHROPIC_API_KEY)

# 4. Define the Editor Agent
@app.post("/frame_classification")
async def classify_frame(
    frame_data: FrameData,
    lr: Annotated[LabelRowV2, Depends(dep_label_row)],
    content: Annotated[NDArray[np.uint8], Depends(dep_single_frame)],
):
    # Receives frame data using FastAPI's Form dependency.
    # Note: FastAPI handles parsing the incoming request body (which implicitly includes frame_data,
    # and the dependencies (dep_label_row, dep_single_frame) resolve the lr and content).

    """Classify a frame using Claude."""
    # Constructs a `Frame` object with the content.
    frame = Frame(frame=frame_data.frame, content=content) 
    
    # Sends the frame image to Claude for analysis.
    message = anthropic_client.messages.create(
        model="claude-3-5-sonnet-20241022",
        max_tokens=1024,
        system=system_prompt,
        messages=[
            {
                "role": "user",
                "content": [frame.b64_encoding(output_format="anthropic")],
            }
        ],
    )
    try:
        # Parses Claude's response into classification instances.
        classifications = data_model(message.content[0].text) 
        for clf in classifications:
            clf.set_for_frames(frame_data.frame, confidence=0.5, manual_annotation=False)
            # Adds classifications to the label row
            lr.add_classification_instance(clf) 
    except Exception:
        import traceback

        traceback.print_exc()
        print(f"Response from model: {message.content[0].text}")

    # Saves the updated data.
    lr.save()
Test the Agent
  1. In your current terminal run the following command to runFastAPI server in development mode with auto-reload enabled.
uvicorn main:app --reload --port 8080
  1. Open your Project in the Encord platform and navigate to a frame you want to add a classification to. Copy the URL from your browser.
The url should have the following format: "https://app.encord.com/label_editor/{project_hash}/{data_hash}/{frame}"
  1. In another shell operating from the same working directory, source your virtual environment and test the agent.
source venv/bin/activate
encord-agents test local frame_classification '<your_url>'
  1. To see if the test is successful, refresh your browser to view the classifications generated by Claude. Once the test runs successfully, you are ready to deploy your agent. Visit the deployment documentation to learn more.

​
Nested Attributes using Claude 3.5 Sonnet

The goals of this example are:
  1. Create an editor agent that can convert generic object annotations (class-less coordinates) into class specific annotations with nested attributes like descriptions, radio buttons, and checklists.
  2. Demonstrate how to use both the OntologyDataModel and the dep_object_crops dependency.
Prerequisites Before you begin, ensure you have: Run the following commands to set up your environment: 
python -m venv venv                 # Create a virtual Python environment  
source venv/bin/activate            # Activate the virtual environment  
python -m pip install encord-agents anthropic  # Install required dependencies  
export ANTHROPIC_API_KEY="<your_api_key>"     # Set your Anthropic API key  
export ENCORD_SSH_KEY_FILE="/path/to/your/private/key"  # Define your Encord access key
Project Setup Create a Project with visual content (images, image groups, image sequences, or videos) in Encord. This example uses the following Ontology, but any Ontology containing classifications can be used provided the object types are the same and there is one entry called “generic”.
ontology.json
{
  "objects": [
    {
      "id": "1",
      "name": "person",
      "color": "#D33115",
      "shape": "bounding_box",
      "featureNodeHash": "2xlDPPAG",
      "required": false,
      "attributes": [
        {
          "id": "1.1",
          "featureNodeHash": "aFCN9MMm",
          "type": "text",
          "name": "activity",
          "required": false,
          "dynamic": false
        }
      ]
    },
    {
      "id": "2",
      "name": "animal",
      "color": "#E27300",
      "shape": "bounding_box",
      "featureNodeHash": "3y6JxTUX",
      "required": false,
      "attributes": [
        {
          "id": "2.1",
          "featureNodeHash": "2P7LTUZA",
          "type": "radio",
          "name": "type",
          "required": false,
          "options": [
            {
              "id": "2.1.1",
              "featureNodeHash": "gJvcEeLl",
              "label": "dolphin",
              "value": "dolphin",
              "options": []
            },
            {
              "id": "2.1.2",
              "featureNodeHash": "CxrftGS4",
              "label": "monkey",
              "value": "monkey",
              "options": []
            },
            {
              "id": "2.1.3",
              "featureNodeHash": "OQyWm7Sm",
              "label": "dog",
              "value": "dog",
              "options": []
            },
            {
              "id": "2.1.4",
              "featureNodeHash": "CDKmYJK/",
              "label": "cat",
              "value": "cat",
              "options": []
            }
          ],
          "dynamic": false
        },
        {
          "id": "2.2",
          "featureNodeHash": "5fFgrM+E",
          "type": "text",
          "name": "description",
          "required": false,
          "dynamic": false
        }
      ]
    },
    {
      "id": "3",
      "name": "vehicle",
      "color": "#16406C",
      "shape": "bounding_box",
      "featureNodeHash": "llw7qdWW",
      "required": false,
      "attributes": [
        {
          "id": "3.1",
          "featureNodeHash": "79mo1G7Q",
          "type": "text",
          "name": "type - short and concise",
          "required": false,
          "dynamic": false
        },
        {
          "id": "3.2",
          "featureNodeHash": "OFrk07Ds",
          "type": "checklist",
          "name": "visible",
          "required": false,
          "options": [
            {
              "id": "3.2.1",
              "featureNodeHash": "KmX/HjRT",
              "label": "wheels",
              "value": "wheels"
            },
            {
              "id": "3.2.2",
              "featureNodeHash": "H6qbEcdj",
              "label": "frame",
              "value": "frame"
            },
            {
              "id": "3.2.3",
              "featureNodeHash": "gZ9OucoQ",
              "label": "chain",
              "value": "chain"
            },
            {
              "id": "3.2.4",
              "featureNodeHash": "cit3aZSz",
              "label": "head lights",
              "value": "head_lights"
            },
            {
              "id": "3.2.5",
              "featureNodeHash": "qQ3PieJ/",
              "label": "tail lights",
              "value": "tail_lights"
            }
          ],
          "dynamic": false
        }
      ]
    },
    {
      "id": "4",
      "name": "generic",
      "color": "#FE9200",
      "shape": "bounding_box",
      "featureNodeHash": "jootTFfQ",
      "required": false,
      "attributes": []
    }
  ],
  "classifications": []
}
To construct the Ontology used in this example, run the following script:
import json
from encord.objects.ontology_structure import OntologyStructure
from encord_agents.core.utils import get_user_client

encord_client = get_user_client()
structure = OntologyStructure.from_dict(json.loads("{the_json_above}"))
ontology = encord_client.create_ontology(
    title="Your ontology title",
    structure=structure
)
print(ontology.ontology_hash)
The goal is to trigger an agent that takes a labeling task from Figure A to Figure B, below: Figure A: No classification labels. Figure B: Multiple nested classification labels generated by an LLM. Create the Agent This section provides the complete code for creating your editor agent, along with an explanation of its internal workings. Agent Setup Steps
  1. Import Dependencies and Configure Project: Import necessary dependencies and set up your project. Remember to insert your project’s unique identifier.
  2. Create a data model and a system prompt based on the Project Ontology to tell Claude how to structure its response.
  3. Initialize Anthropic API Client: Set up an API client to establish communication with the Claude model.
  4. Define the Editor Agent:
  • Arguments are automatically injected when the agent is called (see dependency injection details [suspicious link removed]).
  • The dep_object_crops dependency filters to include only “generic” object crops that still need classification.
  • Call Claude with Image Crops: Use the crop.b64_encoding method to send each image crop to Claude in a format it understands.
  1. Parse Claude’s Response and Update Labels: The data_model parses Claude’s JSON response, creating a new Encord object instance. If successful, the original generic object is replaced with the newly classified instance on the label row.
  2. Save Labels. 
# 1. Import dependencies, authenticate with Encord, and set up the Project.
import os

from anthropic import Anthropic
from encord.objects.ontology_labels_impl import LabelRowV2
from fastapi import Depends
from typing_extensions import Annotated

from encord_agents.core.data_model import InstanceCrop
from encord_agents.core.ontology import OntologyDataModel
from encord_agents.core.utils import get_user_client
from encord_agents.fastapi.cors import get_encord_app
from encord_agents.fastapi.dependencies import (
    FrameData,
    dep_label_row,
    dep_object_crops,
)
# Initialize FastAPI app
app = get_encord_app()

# User client and ontology setup
client = get_user_client()
# Ensure you insert your Project's unique identifier.
project = client.get_project("<project_id>") 
generic_ont_obj, *other_objects = sorted(
    project.ontology_structure.objects,
    key=lambda o: o.title.lower() == "generic",
    reverse=True,
)

# 2. Create a data model and a system prompt based on the Project Ontology to tell Claude how to structure its response.
data_model = OntologyDataModel(other_objects)
system_prompt = f"""
You're a helpful assistant that's supposed to help fill in 
json objects according to this schema:

`{data_model.model_json_schema_str}`

Please only respond with valid json.
"""

# 3. Set up an Anthropic API client to establish communication with the Claude model.
ANTHROPIC_API_KEY = os.getenv("ANTHROPIC_API_KEY")
anthropic_client = Anthropic(api_key=ANTHROPIC_API_KEY)


# 4. Define the Editor Agent.
@app.post("/object_classification")
async def classify_objects(
    frame_data: FrameData,
    lr: Annotated[LabelRowV2, Depends(dep_label_row)],
    crops: Annotated[
        list[InstanceCrop],
        Depends(dep_object_crops(filter_ontology_objects=[generic_ont_obj])),
    ],
):
    """Classify generic objects using Claude."""
    changes = False
    # Iterating through each object crop.
    for crop in crops:
        # 5. Call Claude with Image Crops.
        # Sending each crop image to Claude for analysis.
        message = anthropic_client.messages.create(
            model="claude-3-haiku-20240307",
            max_tokens=1024,
            system=system_prompt,
            messages=[
                {
                    "role": "user",
                    "content": [crop.b64_encoding(output_format="anthropic")],
                }
            ],
        )

        # 6. Parse Claude's Response and Update Labels.
        try:
            # Parsing Claude's response into an updated object instance.
            instance = data_model(message.content[0].text)

            coordinates = crop.instance.get_annotation(frame=frame_data.frame).coordinates
            instance.set_for_frames(
                coordinates=coordinates,
                frames=frame_data.frame,
                confidence=0.5,
                manual_annotation=False,
            )
            # Updating the label row by removing the original object and adding the newly classified instance.
            lr.remove_object(crop.instance) 
            lr.add_object_instance(instance)
            changes = True
        except Exception:
            import traceback

            traceback.print_exc()
            print(f"Response from model: {message.content[0].text}")

    # 7. Save Labels.
    if changes: 
        lr.save()
Testing the Agent
  1. In your current terminal run the following command to runFastAPI server in development mode with auto-reload enabled.
fastapi dev agent.py --port 8080
  1. Open your Project in the Encord platform and navigate to a frame you want to add a classification to. Copy the URL from your browser.
The url should have roughly this format: "https://app.encord.com/label_editor/{project_hash}/{data_hash}/{frame}".
  1. In another shell operating from the same working directory, source your virtual environment and test the agent:
source venv/bin/activate
encord-agents test local object_classification '<your_url>'
  1. To see if the test is successful, refresh your browser to view the classifications generated by Claude. Once the test runs successfully, you are ready to deploy your agent. Visit the deployment documentation to learn more.

​
Video Recaptioning using GPT-4o-mini

The goals of this example are:
  1. Create an Editor Agent that automatically generates multiple variations of video captions.
  2. Demonstrate how to use OpenAI’s GPT-4o-mini model to enhance human-created video captions with a FastAPI-based agent.
Prerequisites Before you begin, ensure you have:
  • Created a virtual Python environment.
  • Installed all necessary dependencies.
  • Have an OpenAI API key.
  • Are able to authenticate with Encord.
Run the following commands to set up your environment: 
python -m venv venv                 # Create a virtual Python environment  
source venv/bin/activate            # Activate the virtual environment  
python -m pip install encord-agents langchain-openai "fastapi[standard]" openai  # Install required dependencies  
export OPENAI_API_KEY="<your-api-key>"     # Set your OpenAI API key  
export ENCORD_SSH_KEY_FILE="/path/to/your/private/key"  # Define your Encord access key
Project Setup Create a Project containing videos in Encord. This example requires an Ontology with four text classifications:
  • One text classification for human-created summaries of what is happening in the video.
  • Three text classifications to be automatically filled by the LLM.
    {
      "objects": [],
      "classifications": [
        {
          "id": "1",
          "featureNodeHash": "GCH8VHIK",
          "attributes": [
            {
              "id": "1.1",
              "name": "Caption",
              "type": "text",
              "required": false,
              "featureNodeHash": "Yg7xXEfC"
            }
          ]
        },
        {
          "id": "2",
          "featureNodeHash": "PwQAwYid",
          "attributes": [
            {
              "id": "2.1",
              "name": "Caption Rephrased 1",
              "type": "text",
              "required": false,
              "featureNodeHash": "aQdXJwbG"
            }
          ]
        },
        {
          "id": "3",
          "featureNodeHash": "3a/aSnHO",
          "attributes": [
            {
              "id": "3.1",
              "name": "Caption Rephrased 2",
              "type": "text",
              "required": false,
              "featureNodeHash": "8zY6H62x"
            }
          ]
        },
        {
          "id": "4",
          "featureNodeHash": "FNjXp5TU",
          "attributes": [
            {
              "id": "4.1",
              "name": "Caption Rephrased 3",
              "type": "text",
              "required": false,
              "featureNodeHash": "sKg1Kq/m"
            }
          ]
        }
      ]
    }
To construct the Ontology used in this example, run the following script:
import json
from encord.objects.ontology_structure import OntologyStructure
from encord.objects.attributes import TextAttribute

structure = OntologyStructure()
caption = structure.add_classification()
caption.add_attribute(TextAttribute, "Caption")
re1 = structure.add_classification()
re1.add_attribute(TextAttribute, "Recaption 1")
re2 = structure.add_classification()
re2.add_attribute(TextAttribute, "Recaption 2")
re3 = structure.add_classification()
re3.add_attribute(TextAttribute, "Recaption 3")

print(json.dumps(structure.to_dict(), indent=2))

create_ontology = False
if create_ontology:
    from encord.user_client import EncordUserClient
    client = EncordUserClient.create_with_ssh_private_key()  # Look in auth section for authentication
    client.create_ontology("title", "description", structure)
The workflow for this agent is:
  1. A human watches the video and enters a caption in the first text field.
  2. The agent is then triggered and generates three additional caption variations for review.
  • Each video is first annotated by a human (ANNOTATE stage).
  • Next, a data agent automatically generates alternative captions (AGENT stage).
  • Finally, a human reviews all four captions (REVIEW stage) before the task is marked complete.
If no human caption is present when the agent is triggered, the task is sent back for annotation. If the review stage results in rejection, the task is also returned for re-annotation.
Create the Agent This section provides the complete code for creating your editor agent, along with an explanation of its internal workings. Agent Setup Steps
  1. Set up imports and create a Pydantic model for our LLM’s structured output
  2. Create a detailed system prompt for the LLM that explains exactly what kind of rephrasing we want
  3. We configure the LLM to use structured outputs based on our model
  4. Create a helper function to prompt the model with both text and image:
  5. Initialize the FastAPI app with the required CORS middleware:
  6. Define the agent to handle the recaptioning. This includes:
  • Retrieving the existing human-created caption, prioritizing captions from the current frame or falling back to frame zero.
  • Sending the first frame of the video along with the human caption to the LLM.
  • Processing the response from the LLM, which provides three alternative phrasings of the original caption.
  • Updating the label row with the new captions, replacing any existing ones.
# 1. Set up imports and create a Pydantic model for our LLM's structured output.
import os
from typing import Annotated

import numpy as np
from encord.exceptions import LabelRowError
from encord.objects.classification_instance import ClassificationInstance
from encord.objects.ontology_labels_impl import LabelRowV2
from fastapi import Depends
from langchain_openai import ChatOpenAI
from numpy.typing import NDArray
from pydantic import BaseModel

from encord_agents import FrameData
from encord_agents.fastapi.cors import get_encord_app
from encord_agents.fastapi.dependencies import Frame, dep_label_row, dep_single_frame

# The response model for the agent to follow.
class AgentCaptionResponse(BaseModel):
    rephrase_1: str
    rephrase_2: str
    rephrase_3: str


# 2. Create a detailed system prompt for the LLM that explains exactly what kind of rephrasing we want.
SYSTEM_PROMPT = """
You are a helpful assistant that rephrases captions.

I will provide you with a video caption and an image of the scene of the video. 

The captions follow this format:

"The droid picks up <cup_0> and puts it on the <table_0>."

The captions that you make should replace the tags, e.g., <cup_0>, with the actual object names.
The replacements should be consistent with the scene.

Here are three rephrases: 

1. The droid picks up the blue mug and puts it on the left side of the table.
2. The droid picks up the cup and puts it to the left of the plate.
3. The droid is picking up the mug on the right side of the table and putting it down next to the plate.

You will rephrase the caption in three different ways, as above, the rephrases should be

1. Diverse in terms of adjectives, object relations, and object positions.
2. Sound in relation to the scene. You cannot talk about objects you cannot see.
3. Short and concise. Keep it within one sentence.

"""

# 3. Configure the LLM to use structured outputs based on our model.
llm = ChatOpenAI(model="gpt-4o-mini", temperature=0.4, api_key=os.environ["OPENAI_API_KEY"])
llm_structured = llm.with_structured_output(AgentCaptionResponse)


# 4. Create a helper function to prompt the model with both text and image.
def prompt_gpt(caption: str, image: Frame) -> AgentCaptionResponse:
    prompt = [
        {"role": "system", "content": SYSTEM_PROMPT},
        {
            "role": "user",
            "content": [
                {"type": "text", "text": f"Video caption: `{caption}`"},
                image.b64_encoding(output_format="openai"),
            ],
        },
    ]
    return llm_structured.invoke(prompt)


# 5. Initialize the FastAPI app with the required CORS middleware.
app = get_encord_app()


# 6. Define the agent to handle the recaptioning.
@app.post("/my_agent")
def my_agent(
    frame_data: FrameData,
    label_row: Annotated[LabelRowV2, Depends(dep_label_row)],
    frame_content: Annotated[NDArray[np.uint8], Depends(dep_single_frame)],
) -> None:
    # Get the relevant Ontology information
    # Recall that we expect
    # [human annotation, llm recaption 1, llm recaption 2, llm recaption 3]
    # in the Ontology
    cap, *rs = label_row.ontology_structure.classifications

    # Retrieve the existing human-created caption, prioritizing captions from the current frame or falling back to frame zero.
    instances = label_row.get_classification_instances(
        filter_ontology_classification=cap, filter_frames=[0, frame_data.frame]
    )
    if not instances:
        # nothing to do if there are no human labels
        return
    elif len(instances) > 1:

        def order_by_current_frame_else_frame_0(
            instance: ClassificationInstance,
        ) -> bool:
            try:
                instance.get_annotation(frame_data.frame)
                return 2  # The best option
            except LabelRowError:
                pass
            try:
                instance.get_annotation(0)
                return 1
            except LabelRowError:
                return 0

        instance = sorted(instances, key=order_by_current_frame_else_frame_0)[-1]
    else:
        instance = instances[0]

    # Read the actual string caption
    caption = instance.get_answer()

    # Send the first frame of the video along with the human caption to the LLM.
    frame = Frame(frame=0, content=frame_content)
    response = prompt_gpt(caption, frame)

    # Process the LLM's response, which contains three different rephrasings of the original caption.
    # Update the label row with the new captions, replacing any existing ones.
    for r, t in zip(rs, [response.rephrase_1, response.rephrase_2, response.rephrase_3]):
        # Overwrite any existing re-captions
        existing_instances = label_row.get_classification_instances(filter_ontology_classification=r)
        for existing_instance in existing_instances:
            label_row.remove_classification(existing_instance)

        # Create new instances
        ins = r.create_instance()
        ins.set_answer(t, attribute=r.attributes[0])
        ins.set_for_frames(0)
        label_row.add_classification_instance(ins)

    label_row.save()
Test the Agent
  1. In your current terminal, run the following command to run the FastAPI server:
ENCORD_SSH_KEY_FILE=/path/to/your_private_key \
OPENAI_API_KEY=<your-api-key> \
fastapi dev main.py
  1. Open your Project in the Encord platform, navigate to a video frame, and add your initial caption. Copy the URL from your browser.
  2. In another shell operating from the same working directory, source your virtual environment and test the agent: 
source venv/bin/activate
encord-agents test local my_agent '<your_url>'
  1. Refresh your browser to view the three AI-generated caption variations. Once the test runs successfully, you are ready to deploy your agent. Visit the deployment documentation to learn more.