Harvest
The Harvest agent connector is a Python package that equips AI agents to interact with Harvest 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.
Connector for the Harvest time-tracking and invoicing API (v2). Provides read access to time tracking data including users, clients, projects, tasks, time entries, invoices, estimates, expenses, and more. Harvest is a cloud-based time tracking and invoicing solution that helps teams track time, manage projects, and streamline invoicing.
Example prompts
The Harvest connector is optimized to handle prompts like these.
- List all users in Harvest
- Show me all active projects
- List all clients
- Show me recent time entries
- List all invoices
- Show me all tasks
- List all expense categories
- Get company information
- How many hours were logged last week?
- Which projects have the most time entries?
- Show me all unbilled time entries
- What are the active projects for a specific client?
- List all overdue invoices
- Which users logged the most hours this month?
Unsupported prompts
The Harvest connector isn't currently able to handle prompts like these.
- Create a new time entry in Harvest
- Update a project budget
- Delete an invoice
- Start a timer for a task
Entities and actions
This connector supports the following entities and actions. For more details, see this connector's full reference documentation.
| Entity | Actions |
|---|---|
| Users | List, Get, Context Store Search |
| Clients | List, Get, Context Store Search |
| Contacts | List, Get, Context Store Search |
| Company | Get, Context Store Search |
| Projects | List, Get, Context Store Search |
| Tasks | List, Get, Context Store Search |
| Time Entries | List, Get, Context Store Search |
| Invoices | List, Get, Context Store Search |
| Invoice Item Categories | List, Get, Context Store Search |
| Estimates | List, Get, Context Store Search |
| Estimate Item Categories | List, Get, Context Store Search |
| Expenses | List, Get, Context Store Search |
| Expense Categories | List, Get, Context Store Search |
| Roles | List, Get, Context Store Search |
| User Assignments | List, Context Store Search |
| Task Assignments | List, Context Store Search |
| Time Projects | List, Context Store Search |
| Time Tasks | List, Context Store Search |
Harvest API docs
See the official Harvest API reference.
SDK installation
uv pip install airbyte-agent-sdk
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 HarvestConnector and reads AIRBYTE_CLIENT_ID / AIRBYTE_CLIENT_SECRET from the environment:
- Pydantic AI
- LangChain
- OpenAI Agents
- FastMCP
Pydantic AI
from pydantic_ai import Agent
from airbyte_agent_sdk import connect
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
connector = connect("harvest", workspace_name="<your_workspace_name>")
agent = Agent("openai:gpt-4o")
@agent.tool_plain
@HarvestConnector.tool_utils
async def harvest_execute(entity: str, action: str, params: dict | None = None):
return await connector.execute(entity, action, params or {})
LangChain
from langchain_core.tools import tool
from airbyte_agent_sdk import connect
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
connector = connect("harvest", workspace_name="<your_workspace_name>")
@tool
@HarvestConnector.tool_utils
async def harvest_execute(entity: str, action: str, params: dict | None = None):
"""Execute Harvest connector operations."""
result = await connector.execute(entity, action, params or {})
# connector.execute returns a Pydantic envelope for typed actions; fall back to raw data otherwise.
return result.model_dump(mode="json") if hasattr(result, "model_dump") else result
OpenAI Agents
from agents import Agent, function_tool
from airbyte_agent_sdk import connect
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
connector = connect("harvest", workspace_name="<your_workspace_name>")
# strict_mode=False because `params: dict` is permissive and the default strict
# JSON schema rejects objects with additionalProperties.
@function_tool(strict_mode=False)
@HarvestConnector.tool_utils(framework="openai_agents")
async def harvest_execute(entity: str, action: str, params: dict | None = None):
"""Execute Harvest connector operations."""
result = await connector.execute(entity, action, params or {})
return result.model_dump(mode="json") if hasattr(result, "model_dump") else result
agent = Agent(name="Harvest Assistant", tools=[harvest_execute])
FastMCP
from fastmcp import FastMCP
from airbyte_agent_sdk import connect
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
connector = connect("harvest", workspace_name="<your_workspace_name>")
mcp = FastMCP("Harvest Agent")
@mcp.tool
@HarvestConnector.tool_utils
async def harvest_execute(entity: str, action: str, params: dict | None = None):
"""Execute Harvest connector operations."""
result = await connector.execute(entity, action, params or {})
return result.model_dump(mode="json") if hasattr(result, "model_dump") else result
Or pass credentials explicitly (equivalent, useful when you're not loading them from the environment):
- Pydantic AI
- LangChain
- OpenAI Agents
- FastMCP
Pydantic AI
from pydantic_ai import Agent
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
from airbyte_agent_sdk.types import AirbyteAuthConfig
connector = HarvestConnector(
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>"
)
)
agent = Agent("openai:gpt-4o")
@agent.tool_plain
@HarvestConnector.tool_utils
async def harvest_execute(entity: str, action: str, params: dict | None = None):
return await connector.execute(entity, action, params or {})
LangChain
from langchain_core.tools import tool
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
from airbyte_agent_sdk.types import AirbyteAuthConfig
connector = HarvestConnector(
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>"
)
)
@tool
@HarvestConnector.tool_utils
async def harvest_execute(entity: str, action: str, params: dict | None = None):
"""Execute Harvest connector operations."""
result = await connector.execute(entity, action, params or {})
# connector.execute returns a Pydantic envelope for typed actions; fall back to raw data otherwise.
return result.model_dump(mode="json") if hasattr(result, "model_dump") else result
OpenAI Agents
from agents import Agent, function_tool
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
from airbyte_agent_sdk.types import AirbyteAuthConfig
connector = HarvestConnector(
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>"
)
)
# strict_mode=False because `params: dict` is permissive and the default strict
# JSON schema rejects objects with additionalProperties.
@function_tool(strict_mode=False)
@HarvestConnector.tool_utils(framework="openai_agents")
async def harvest_execute(entity: str, action: str, params: dict | None = None):
"""Execute Harvest connector operations."""
result = await connector.execute(entity, action, params or {})
return result.model_dump(mode="json") if hasattr(result, "model_dump") else result
agent = Agent(name="Harvest Assistant", tools=[harvest_execute])
FastMCP
from fastmcp import FastMCP
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
from airbyte_agent_sdk.types import AirbyteAuthConfig
connector = HarvestConnector(
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>"
)
)
mcp = FastMCP("Harvest Agent")
@mcp.tool
@HarvestConnector.tool_utils
async def harvest_execute(entity: str, action: str, params: dict | None = None):
"""Execute Harvest connector operations."""
result = await connector.execute(entity, action, params or {})
return result.model_dump(mode="json") if hasattr(result, "model_dump") else result
Open source
In open source mode, you provide API credentials directly to the connector.
- Pydantic AI
- LangChain
- OpenAI Agents
- FastMCP
Pydantic AI
from pydantic_ai import Agent
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
from airbyte_agent_sdk.connectors.harvest.models import HarvestPersonalAccessTokenAuthConfig
connector = HarvestConnector(
auth_config=HarvestPersonalAccessTokenAuthConfig(
token="<Your Harvest personal access token>",
account_id="<Your Harvest account ID>"
)
)
agent = Agent("openai:gpt-4o")
@agent.tool_plain
@HarvestConnector.tool_utils
async def harvest_execute(entity: str, action: str, params: dict | None = None):
return await connector.execute(entity, action, params or {})
LangChain
from langchain_core.tools import tool
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
from airbyte_agent_sdk.connectors.harvest.models import HarvestPersonalAccessTokenAuthConfig
connector = HarvestConnector(
auth_config=HarvestPersonalAccessTokenAuthConfig(
token="<Your Harvest personal access token>",
account_id="<Your Harvest account ID>"
)
)
@tool
@HarvestConnector.tool_utils
async def harvest_execute(entity: str, action: str, params: dict | None = None):
"""Execute Harvest connector operations."""
result = await connector.execute(entity, action, params or {})
# connector.execute returns a Pydantic envelope for typed actions; fall back to raw data otherwise.
return result.model_dump(mode="json") if hasattr(result, "model_dump") else result
OpenAI Agents
from agents import Agent, function_tool
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
from airbyte_agent_sdk.connectors.harvest.models import HarvestPersonalAccessTokenAuthConfig
connector = HarvestConnector(
auth_config=HarvestPersonalAccessTokenAuthConfig(
token="<Your Harvest personal access token>",
account_id="<Your Harvest account ID>"
)
)
# strict_mode=False because `params: dict` is permissive and the default strict
# JSON schema rejects objects with additionalProperties.
@function_tool(strict_mode=False)
@HarvestConnector.tool_utils(framework="openai_agents")
async def harvest_execute(entity: str, action: str, params: dict | None = None):
"""Execute Harvest connector operations."""
result = await connector.execute(entity, action, params or {})
return result.model_dump(mode="json") if hasattr(result, "model_dump") else result
agent = Agent(name="Harvest Assistant", tools=[harvest_execute])
FastMCP
from fastmcp import FastMCP
from airbyte_agent_sdk.connectors.harvest import HarvestConnector
from airbyte_agent_sdk.connectors.harvest.models import HarvestPersonalAccessTokenAuthConfig
connector = HarvestConnector(
auth_config=HarvestPersonalAccessTokenAuthConfig(
token="<Your Harvest personal access token>",
account_id="<Your Harvest account ID>"
)
)
mcp = FastMCP("Harvest Agent")
@mcp.tool
@HarvestConnector.tool_utils
async def harvest_execute(entity: str, action: str, params: dict | None = None):
"""Execute Harvest connector operations."""
result = await connector.execute(entity, action, params or {})
return result.model_dump(mode="json") if hasattr(result, "model_dump") else result
Authentication
For all authentication options, see the connector's authentication documentation.
Version information
Connector version: 1.0.5