Skip to main content

Facebook-Marketing

The Facebook-Marketing agent connector is a Python package that equips AI agents to interact with Facebook-Marketing through strongly typed, well-documented tools. It's ready to use directly in your Python app, in an agent framework, or exposed through an MCP.

Facebook Marketing API connector for managing ad campaigns, ad sets, ads, creatives, and accessing performance insights, pixel configuration, and event quality data. This connector provides read access to Facebook Ads Manager data for analytics and reporting purposes.

Example prompts​

The Facebook-Marketing connector is optimized to handle prompts like these.

  • List all active campaigns in my ad account
  • What ads are currently running in a recent campaign?
  • List all ad creatives in my account
  • What is the status of my campaigns?
  • List all custom conversion events in my account
  • Show me all ad images in my account
  • What videos are available in my ad account?
  • Create a new campaign called 'Summer Sale 2026' with traffic objective
  • Pause my most recent campaign
  • Create a new ad set with a $50 daily budget in my latest campaign
  • Update the daily budget of my top performing ad set to $100
  • Rename my most recent ad set to 'Holiday Promo'
  • Create a new ad in my latest ad set
  • Pause all ads in my most recent ad set
  • List all pixels in my ad account
  • Show me the event stats for my pixel
  • What events is my Facebook pixel tracking?
  • Search the Ad Library for political ads in the US
  • Find ads about climate change in the Ad Library
  • Show me Ad Library ads from a specific Facebook page
  • Show me the ad sets with the highest daily budget
  • Show me the performance insights for the last 7 days
  • Which campaigns have the most spend this month?
  • Show me ads with the highest click-through rate

Unsupported prompts​

The Facebook-Marketing connector isn't currently able to handle prompts like these.

  • Delete this ad creative
  • Delete this campaign
  • Delete this ad set
  • Delete this ad

Entities and actions​

This connector supports the following entities and actions. For more details, see this connector's full reference documentation.

EntityActions
Current UserGet
Ad AccountsList, Get, Context Store Search, Context Store SQL Query
CampaignsList, Create, Get, Update, Context Store Search, Context Store SQL Query
Ad SetsList, Create, Get, Update, Context Store Search, Context Store SQL Query
AdsList, Create, Get, Update, Context Store Search, Context Store SQL Query
Ad CreativesList, Context Store Search, Context Store SQL Query, Semantic Search
Ads InsightsList, Context Store Search, Context Store SQL Query
Custom ConversionsList, Context Store Search, Context Store SQL Query
ImagesList, Context Store Search, Context Store SQL Query
VideosList, Context Store Search, Context Store SQL Query
PixelsList, Get
Pixel StatsList
Ad LibraryList

Facebook-Marketing API docs​

See the official Facebook-Marketing API reference.

Interfaces​

Use the Facebook-Marketing connector through the Airbyte Agent CLI, the Python SDK, or the API.

CLI​

Install the CLI:

curl -fsSL https://airbyte.ai/install.sh | bash

Authenticate with Airbyte:

airbyte-agent login

Create the connector. The CLI opens the hosted setup flow:

airbyte-agent connectors create --json '{
  "workspace": "<your_workspace_name>",
  "name": "facebook-marketing"
}'

Describe the connector to see its supported entities and actions:

airbyte-agent connectors describe --json '{
  "workspace": "<your_workspace_name>",
  "name": "facebook-marketing"
}'

Execute an action:

airbyte-agent connectors execute --json '{
  "workspace": "<your_workspace_name>",
  "name": "facebook-marketing",
  "entity": "current_user",
  "action": "get"
}'

Python SDK​

Installation​

uv pip install airbyte-agent-sdk

Usage​

Connectors can run in hosted or open source mode.

Hosted​

In hosted mode, API credentials are stored securely in Airbyte Agents. You provide your Airbyte credentials instead. If your Airbyte client can access multiple organizations, also set organization_id.

This example assumes you've already authenticated your connector with Airbyte. See Authentication to learn more about authenticating. If you need a step-by-step guide, see the hosted execution tutorial.

The connect() factory returns a fully typed FacebookMarketingConnector and reads AIRBYTE_CLIENT_ID / AIRBYTE_CLIENT_SECRET from the environment:

The recommended pattern is build_connector_tools, which gives the agent three tools bound to this connector: inspect_connector, read_skill_docs, and execute. The agent can inspect the connector, read only the skill-doc section it needs, and then execute:

inspect_connector() -> read_skill_docs() -> read_skill_docs(section="...") -> execute(entity, action, params)

Pass section IDs verbatim as the outline lists them, prefix included (actions.<entity>.<action>, not <entity>.<action>); anything else returns an error the agent has to recover from.

The builder names its tools inspect_connector, read_skill_docs, and execute, so the tool sets for more than one connector collide when registered on the same agent. Renaming the callables at registration avoids the collision, but the generated execute guidance still names inspect_connector and read_skill_docs, pointing the model at the wrong tools. Use the agent_tool pattern below instead: it weaves your own names into that guidance.

Pydantic AI
from airbyte_agent_sdk import build_connector_tools
from pydantic_ai import Agent
from airbyte_agent_sdk import connect
from airbyte_agent_sdk.connectors.facebook_marketing import FacebookMarketingConnector

connector = connect("facebook-marketing", workspace_name="<your_workspace_name>")

tools = build_connector_tools(connector, framework="pydantic_ai")
agent = Agent("openai:gpt-4o", tools=tools.as_list())
Custom tool bodies​

When you need custom tool bodies — or a framework without native support — use FacebookMarketingConnector.agent_tool. Register execute, inspect, and docs together so the agent can fetch connector guidance progressively. Pass the framework explicitly when it has a supported failure strategy:

Pydantic AI
from pydantic_ai import Agent
from airbyte_agent_sdk import connect
from airbyte_agent_sdk.connectors.facebook_marketing import FacebookMarketingConnector

connector = connect("facebook-marketing", workspace_name="<your_workspace_name>")

agent = Agent("openai:gpt-4o")

@agent.tool_plain
@FacebookMarketingConnector.agent_tool(
    framework="pydantic_ai",
    inspect_tool="facebook_marketing_inspect",
    docs_tool="facebook_marketing_read_docs",
)
async def facebook_marketing_execute(entity: str, action: str, params: dict | None = None):
    return await connector.execute(entity, action, params or {})

@agent.tool_plain
@FacebookMarketingConnector.agent_tool(framework="pydantic_ai")
async def facebook_marketing_inspect():
    return await connector.inspect_connector()

@agent.tool_plain
@FacebookMarketingConnector.agent_tool(framework="pydantic_ai")
async def facebook_marketing_read_docs(section: str | None = None):
    return await connector.read_skill_docs(section)

Use the same three-function pattern with framework="langchain", "openai_agents", or "mcp" and that framework's registration decorator. Each value translates connector failures into the framework's own signal:

framework=Tool failures surface as
"pydantic_ai"pydantic_ai.ModelRetry
"langchain"langchain_core.tools.ToolException (set handle_tool_error=True to feed it back to the model)
"openai_agents"the failure message returned to the model as the tool result
"mcp"fastmcp.exceptions.ToolError
"none" (default)airbyte_agent_sdk.AirbyteToolError

On a framework the SDK does not support natively — or in a raw LLM dispatch loop — omit framework= and handle AirbyteToolError yourself:

No framework
from airbyte_agent_sdk import AirbyteToolError
from airbyte_agent_sdk import connect
from airbyte_agent_sdk.connectors.facebook_marketing import FacebookMarketingConnector

connector = connect("facebook-marketing", workspace_name="<your_workspace_name>")

@FacebookMarketingConnector.agent_tool(
    inspect_tool="facebook_marketing_inspect",
    docs_tool="facebook_marketing_read_docs",
)
async def facebook_marketing_execute(entity: str, action: str, params: dict | None = None):
    return await connector.execute(entity, action, params or {})

@FacebookMarketingConnector.agent_tool()
async def facebook_marketing_inspect():
    return await connector.inspect_connector()

@FacebookMarketingConnector.agent_tool()
async def facebook_marketing_read_docs(section: str | None = None):
    return await connector.read_skill_docs(section)

# Advertise all three to the model, using each function's docstring as its description.
handlers = {
    fn.__name__: fn
    for fn in (facebook_marketing_inspect, facebook_marketing_read_docs, facebook_marketing_execute)
}

# `tool_name` and `tool_args` come from the model's tool call in your dispatch loop.
try:
    tool_result = await handlers[tool_name](**tool_args)
except AirbyteToolError as err:
    tool_result = str(err)  # hand the message back to the model as an errored tool result

Each function's docstring carries the guidance the model needs, so pass it through as the tool description wherever you register it.

Legacy alternatives​

These examples are kept for existing integrations. The deprecated FacebookMarketingConnector.tool_utils pattern loads the connector's full generated catalog into one broad execute tool description instead of letting the agent read skill docs on demand. For new code, use build_connector_tools or FacebookMarketingConnector.agent_tool above.

Pydantic AI
from pydantic_ai import Agent
from airbyte_agent_sdk import connect
from airbyte_agent_sdk.connectors.facebook_marketing import FacebookMarketingConnector

connector = connect("facebook-marketing", workspace_name="<your_workspace_name>")

agent = Agent("openai:gpt-4o")

@agent.tool_plain
@FacebookMarketingConnector.tool_utils
async def facebook_marketing_execute(entity: str, action: str, params: dict | None = None):
    return await connector.execute(entity, action, params or {})

Or pass credentials explicitly (equivalent, useful when you're not loading them from the environment):

Pydantic AI
from airbyte_agent_sdk import build_connector_tools
from pydantic_ai import Agent
from airbyte_agent_sdk.connectors.facebook_marketing import FacebookMarketingConnector
from airbyte_agent_sdk.types import AirbyteAuthConfig

connector = FacebookMarketingConnector(
    auth_config=AirbyteAuthConfig(
        workspace_name="<your_workspace_name>",
        organization_id="<your_organization_id>",  # Optional for multi-org clients
        airbyte_client_id="<your-client-id>",
        airbyte_client_secret="<your-client-secret>"
    )
)

tools = build_connector_tools(connector, framework="pydantic_ai")
agent = Agent("openai:gpt-4o", tools=tools.as_list())
Open source​

In open source mode, you provide API credentials directly to the connector.

The recommended pattern is build_connector_tools, which gives the agent three tools bound to this connector: inspect_connector, read_skill_docs, and execute. The agent can inspect the connector, read only the skill-doc section it needs, and then execute:

inspect_connector() -> read_skill_docs() -> read_skill_docs(section="...") -> execute(entity, action, params)

Pass section IDs verbatim as the outline lists them, prefix included (actions.<entity>.<action>, not <entity>.<action>); anything else returns an error the agent has to recover from.

The builder names its tools inspect_connector, read_skill_docs, and execute, so the tool sets for more than one connector collide when registered on the same agent. Renaming the callables at registration avoids the collision, but the generated execute guidance still names inspect_connector and read_skill_docs, pointing the model at the wrong tools. Use the agent_tool pattern below instead: it weaves your own names into that guidance.

Pydantic AI
from airbyte_agent_sdk import build_connector_tools
from pydantic_ai import Agent
from airbyte_agent_sdk.connectors.facebook_marketing import FacebookMarketingConnector
from airbyte_agent_sdk.connectors.facebook_marketing.models import FacebookMarketingServiceAccountKeyAuthenticationAuthConfig

connector = FacebookMarketingConnector(
    auth_config=FacebookMarketingServiceAccountKeyAuthenticationAuthConfig(
        account_key="<Facebook long-lived access token for Service Account authentication>"
    )
)

tools = build_connector_tools(connector, framework="pydantic_ai")
agent = Agent("openai:gpt-4o", tools=tools.as_list())
Custom tool bodies​

When you need custom tool bodies — or a framework without native support — use FacebookMarketingConnector.agent_tool. Register execute, inspect, and docs together so the agent can fetch connector guidance progressively. Pass the framework explicitly when it has a supported failure strategy:

Pydantic AI
from pydantic_ai import Agent
from airbyte_agent_sdk.connectors.facebook_marketing import FacebookMarketingConnector
from airbyte_agent_sdk.connectors.facebook_marketing.models import FacebookMarketingServiceAccountKeyAuthenticationAuthConfig

connector = FacebookMarketingConnector(
    auth_config=FacebookMarketingServiceAccountKeyAuthenticationAuthConfig(
        account_key="<Facebook long-lived access token for Service Account authentication>"
    )
)

agent = Agent("openai:gpt-4o")

@agent.tool_plain
@FacebookMarketingConnector.agent_tool(
    framework="pydantic_ai",
    inspect_tool="facebook_marketing_inspect",
    docs_tool="facebook_marketing_read_docs",
)
async def facebook_marketing_execute(entity: str, action: str, params: dict | None = None):
    return await connector.execute(entity, action, params or {})

@agent.tool_plain
@FacebookMarketingConnector.agent_tool(framework="pydantic_ai")
async def facebook_marketing_inspect():
    return await connector.inspect_connector()

@agent.tool_plain
@FacebookMarketingConnector.agent_tool(framework="pydantic_ai")
async def facebook_marketing_read_docs(section: str | None = None):
    return await connector.read_skill_docs(section)

Use the same three-function pattern with framework="langchain", "openai_agents", or "mcp" and that framework's registration decorator. Each value translates connector failures into the framework's own signal:

framework=Tool failures surface as
"pydantic_ai"pydantic_ai.ModelRetry
"langchain"langchain_core.tools.ToolException (set handle_tool_error=True to feed it back to the model)
"openai_agents"the failure message returned to the model as the tool result
"mcp"fastmcp.exceptions.ToolError
"none" (default)airbyte_agent_sdk.AirbyteToolError

On a framework the SDK does not support natively — or in a raw LLM dispatch loop — omit framework= and handle AirbyteToolError yourself:

No framework
from airbyte_agent_sdk import AirbyteToolError
from airbyte_agent_sdk.connectors.facebook_marketing import FacebookMarketingConnector
from airbyte_agent_sdk.connectors.facebook_marketing.models import FacebookMarketingServiceAccountKeyAuthenticationAuthConfig

connector = FacebookMarketingConnector(
    auth_config=FacebookMarketingServiceAccountKeyAuthenticationAuthConfig(
        account_key="<Facebook long-lived access token for Service Account authentication>"
    )
)

@FacebookMarketingConnector.agent_tool(
    inspect_tool="facebook_marketing_inspect",
    docs_tool="facebook_marketing_read_docs",
)
async def facebook_marketing_execute(entity: str, action: str, params: dict | None = None):
    return await connector.execute(entity, action, params or {})

@FacebookMarketingConnector.agent_tool()
async def facebook_marketing_inspect():
    return await connector.inspect_connector()

@FacebookMarketingConnector.agent_tool()
async def facebook_marketing_read_docs(section: str | None = None):
    return await connector.read_skill_docs(section)

# Advertise all three to the model, using each function's docstring as its description.
handlers = {
    fn.__name__: fn
    for fn in (facebook_marketing_inspect, facebook_marketing_read_docs, facebook_marketing_execute)
}

# `tool_name` and `tool_args` come from the model's tool call in your dispatch loop.
try:
    tool_result = await handlers[tool_name](**tool_args)
except AirbyteToolError as err:
    tool_result = str(err)  # hand the message back to the model as an errored tool result

Each function's docstring carries the guidance the model needs, so pass it through as the tool description wherever you register it.

Legacy alternatives​

These examples are kept for existing integrations. The deprecated FacebookMarketingConnector.tool_utils pattern loads the connector's full generated catalog into one broad execute tool description instead of letting the agent read skill docs on demand. For new code, use build_connector_tools or FacebookMarketingConnector.agent_tool above.

Pydantic AI
from pydantic_ai import Agent
from airbyte_agent_sdk.connectors.facebook_marketing import FacebookMarketingConnector
from airbyte_agent_sdk.connectors.facebook_marketing.models import FacebookMarketingServiceAccountKeyAuthenticationAuthConfig

connector = FacebookMarketingConnector(
    auth_config=FacebookMarketingServiceAccountKeyAuthenticationAuthConfig(
        account_key="<Facebook long-lived access token for Service Account authentication>"
    )
)

agent = Agent("openai:gpt-4o")

@agent.tool_plain
@FacebookMarketingConnector.tool_utils
async def facebook_marketing_execute(entity: str, action: str, params: dict | None = None):
    return await connector.execute(entity, action, params or {})

Authentication​

For all authentication options, see the connector's authentication documentation.

IP allow list​

If your organization restricts access to specific IPs, add the Airbyte Agents IP addresses to your allow list.

Version information​

Connector version: 1.2.0

Was this page helpful?