AI agents that can understand and refactor code become truly powerful development companions. Whether it's finding function definitions, renaming variables across files, or extracting code into reusable functions, Toolpack SDK's coding tools give your agents the ability to navigate and modify JavaScript/TypeScript codebases with AST-level precision.

In this post, we'll explore all 12 tools in the coding category — from symbol navigation to multi-file refactoring with atomic edits.

AI agents that can interact with APIs and download files become truly powerful integration tools. Whether it's fetching data from REST endpoints, sending webhooks, or downloading resources, Toolpack SDK's HTTP tools give your agents the ability to communicate with external services seamlessly.

In this post, we'll explore all 5 tools in the network category — from simple GET requests to file downloads with custom headers.

AI agents that can run commands and manage processes become truly powerful automation tools. Whether it's running tests, starting development servers, or executing deployment scripts, Toolpack SDK's execution tools give your agents the ability to interact with the system shell safely and effectively.

In this post, we'll explore all 6 tools in the execution category — from simple command execution to background process management with full lifecycle control.

AI agents that modify files or run commands need to know where they are. What OS they're on. How much disk is left. What env vars are set. Without that context, you get an agent that assumes Linux paths on Windows or writes a 2GB file to a disk with 500MB free.

Toolpack SDK's system tools give your agents the context they need to operate safely and effectively across different environments.

Toolpack SDK's 18 filesystem tools are powerful on their own, but their real potential emerges when you combine them into patterns that solve real-world problems. This guide explores advanced techniques for building robust, efficient, and production-ready AI agents that work with files.

We'll cover atomic operations, intelligent file discovery, batch processing strategies, error handling patterns, and complete examples of production-grade filesystem workflows.

The Toolpack SDK workflow engine is powerful out of the box, but its real strength emerges when you customize it. Instead of relying solely on the built-in AGENT_WORKFLOW, CODING_WORKFLOW, or CHAT_WORKFLOW presets, you can craft workflows that match your exact requirements—whether that's a research agent, a code reviewer, or a deployment orchestrator.

This guide shows you how to build custom workflows from scratch, tune planning behavior, control step execution, and design workflows that fit your specific use cases.

The @toolpack-sdk/knowledge package provides Retrieval-Augmented Generation (RAG) capabilities for your AI agents. Build knowledge bases from documentation, code, or any text source and enable semantic search within your agent conversations. This post covers everything you need to know about creating knowledge bases, configuring providers and embedders, and integrating with the Toolpack SDK.

An AI agent that can read files, run shell commands, and hit external APIs is useful. An AI agent that does all of that when you only asked it to answer a question is a problem.

Modes are how Toolpack SDK draws that line. A mode defines three things: which tools the AI can access, what system prompt shapes its behavior, and whether the workflow engine (planning, step execution, retries) is active. You set the mode once at init or switch it at runtime, and the SDK enforces the constraints on every request.

Toolpack SDK supports four built-in AI providers out of the box: OpenAI, Anthropic, Gemini, and Ollama. You use your own API keys directly — no subscriptions, no middleman. This post covers everything you need to know about configuring providers, switching between them at runtime, and listing available models.

Toolpack SDK ships with a workflow engine that turns a basic AI chat into a goal-driven agent. Instead of responding once and waiting, it plans a series of steps, executes them one by one, handles failures, and reports progress the whole time.

This guide walks you through both workflow types, how to configure them, how to listen to events, and patterns you can use in real applications.