What are Tool Nodes and who should use them?
Tool Nodes are purpose-built connections within the Hoop platform that allow your agent to interact with external services, knowledge bases, and advanced reasoning models like Large Language Models (LLMs). They are designed for anyone building sophisticated agents that need to go beyond basic conversation. This includes operations teams, product managers, and developers looking to automate complex workflows and ensure agents can fetch real-time data or take action in third-party systems.
Tool Nodes are essential for creating agents that are truly capable of multi-step, intelligent tasks, making your Hoop-powered solutions more powerful and versatile across various business functions.
How do Tool Nodes benefit my agent's capabilities?
Tool Nodes dramatically extend your agent's functionality by integrating LLMs, knowledge, APIs, and web search directly into its workflow. This means your agent can reason over current data, access proprietary information, and execute actions in other applications. The immediate benefit is moving from simple replies to complex task completion.
Key benefits include:
- Enabling agents to perform multi-step tasks confidently.
- Maintaining secure and auditable connections to external systems.
- Providing access to up-to-date information via web search or internal databases.
- Expanding the types of problems your automated agents can solve.
What types of integrations are possible with Tool Nodes?
Tool Nodes support a broad spectrum of integrations, allowing you to connect to virtually any external system. Standardized connectors make it easy to reach external systems with clear scopes and auditability, ensuring operations are safe and secure within the Hoop environment. The currently available node types focus on core intelligent and operational needs.
You can connect your agent using these available node types:
- LLM (Large Language Model) Nodes for advanced prompting.
- API Tool Nodes for connecting to any external web service.
- Knowledge Base Nodes for accessing internal data.
- Web Search Nodes for real-time information retrieval.
- MCP Tool Nodes for platform-specific capabilities.
How is security and data access handled within Tool Nodes?
Security is paramount when connecting your Hoop agents to external systems. Each Tool Node explicitly declares the data scopes and permissions it requires. Administrators have full control over connection approvals and can monitor usage via a comprehensive audit trail, ensuring transparency and compliance. This model protects sensitive information and prevents unauthorized access or actions.
When using APIs, you must provide the necessary credentials (such as API keys) within secure connections managed by the platform. The node references these credentials only at runtime, never exposing them in the general workflow configuration. This strict control over credentials maintains the security integrity of your external systems.
What is the pricing and plan availability for Tool Nodes?
Tool Nodes are a core component of powerful agents and are available across multiple tiers of the Hoop platform. The usage limits and specific features related to complex chaining and advanced nodes scale with your subscription level. This ensures that businesses of all sizes can leverage intelligent automation.
The feature is available in the following plans:
Higher-tier plans typically offer greater capacity for node execution, more specialized node types, and enhanced administrative features for managing large-scale deployments and audit logs.
What steps are required to set up and configure a new Tool Node?
Setting up a Tool Node is designed to be fast and intuitive within the Hoop interface. You simply drop the appropriate node—such as an API Tool or a Knowledge Base Node—into your agent's workflow. Configuration is completed in minutes, requiring you to define the necessary parameters and securely provide credentials if connecting to an external API.
This ease of configuration, combined with clear scopes and standardized connectors, means you can quickly extend your agent's functionality and see value almost immediately. Once connected and configured, the node is instantly available for your agent's reasoning process.
.avif)
%20(1).avif){kind=avatar}
.avif){kind=avatar}
