An AI tool is a skill or capability that the Thunk.AI platform gives an AI agent to extend what it can do, to bring in information it doesn't have, or for it to record its work. Instead of responding to an instruction with a message, an AI agent can respond by invoking an AI tool with appropriate parameters. You can think of a tool invocation as a structured response from the AI agent. Thunk.AI makes extensive use of tool invocations as the primary mechanism for AI agents to respond.
AI tools are used to connect an AI agent to the rest of the business work environment and the other applications and data in that environment. Some tools are used to fetch information (eg: search a company database) while other tools perform actions (eg: create new documents or update external systems).
Tool Libraries
Most designers of thunks do not have to build AI tools from scratch. Many AI tools are built into the system in functionality bundles called AI Tool Libraries. There are different kinds of libraries:
Standard libraries: these are a default set of AI tool libraries with capabilities like web search and browsing, email drafting, image and video processing, and document processing.
Web search: tools to search the web, answer questions based on web content, read and extract information from web pages
File system module: tools to read, create, update, move, and delete files in a cloud file system (Google Drive or Office 365, depending on the user's work environment).
Image processing: tools to edit and query image content
Video processing: tools to edit and query video content
Communications: tools to draft or send email and messages
Document templating: tools to create documents based on parameterized templates
Application libraries: every user can augment their account with connections to other systems. For example, a user may add a Google Drive connection and this automatically enables an AI tool library that allows that user's AI agents to read and write files and spreadsheets and folders in that Google Drive. Another common use of connections is to integrate with an existing enterprise application using the MCP protocol. Learn more about integration with applications.
Imported thunk libraries: users can define AI tools in one thunk and export them to share and reuse across other thunks (of their own or for access to other users as well). There are many benefits to such modular reuse, especially when teams of users are implementing many AI agent automation processes. Learn more about modular reuse of thunks.
Platform libraries: some of the internal capabilities of the platform (eg: dynamic planning, or reflection, or workflow state update) are also represented as AI tools. While the user cannot directly author these tools, they can be configured in the same way as any other AI tool.
Custom "This-Thunk" Tools
In each thunk, the owner/admin (with the help of their AI agent) can also create custom AI tools (eg: to connect to specific enterprise systems or databases). There are four categories of custom AI tools: API tools, database tools, code-based tools, and intelligent tools.
REST API Tools
API tools that can invoke any REST API using HTTPS. The API tools can utilize connection credentials that have been saved at the account level. This ensures that connection credentials do not have to be repeated in every tool.
API tools have completely deterministic behavior and act as an interface between the intelligent logic of a thunk and traditional software systems.
Database Tools
Database query and update tools that invoke any SQL commands with a database connection. A variety of database connection types are supported.
Database tools have completely deterministic behavior and act as an interface between the intelligent logic of a thunk and traditional database systems which hold a lot of enterprise data and context.
Code-based Tools
Code-based tools that implement logic using deterministic Typescript code. This code is hosted in a safe sandbox. Such code-based tools are useful for special-case modular logic that requires the complexity of traditional code. Often, these tools are used to parse a complex data payload coming from another system.
Intelligent AI-based Tools
Intelligent tools combine natural language instructions with existing tools to build nested AI logic wrapped into the form of a tool. This is a powerful and modular way to construct AI applications. Each intelligent tool has its own AI instructions (prompts, tool configuration, etc) and runs as an independent AI agent on each invocation. Since the same tool can be used in different steps of a thunk, this becomes a mechanism to define and reuse modular AI logic within a thunk. And also, cince tools can be exported from the thunk in the form of an exported tool library, it adds greater power to reuse modular AI logic across multiple thunks.
Content-based Tools
Each of the content folders in a thunk is automatically associated with a search tool that allows an AI agent to search the content and find relevant information.
Tool Configuration
It is important that an AI agent gets the AI tools it needs, but it is also important that unnecessary tools are not provided. These can add cost, delay, and even cause unreliable agentic behavior.
AI tools can be configured at the level of the thunk as a whole. Entire tool libraries can be enabled or disabled for the thunk, and individual tools within each enabled library can be further disabled.
Within every AI instruction at the level of a workflow step, the tool libraries and tools enabled at the thunk level are automatically enabled. They can be further configured at this level to disable some of the libraries and tools.
Enabled tools can also be constrained via natural language instructions. For example, a web search tool can be constrained within one step with a constraint like this: "only search within site fda.gov".