Tendril
Loading lesson…
Agents are only as useful as their tools. Tour the big three — filesystem, browser, code execution — plus the emerging MCP ecosystem, with examples of what each unlocks.
A language model without tools can only talk. Add a browser and it can research. Add a filesystem and it can build. Add code execution and it can compute. Tools turn 'smart autocomplete' into 'team member who can actually do things.'
| Tool | What it unlocks | 2026 reference products |
|---|---|---|
| Filesystem | Read, write, move, delete files. | Claude Code, Devin, OpenClaw. |
| Browser | Open pages, click, fill forms, scrape. | Computer Use, Browser Use, Operator, Atlas. |
| Code execution | Run scripts in Python, JS, shell. | Code Interpreter (ChatGPT), Claude Analysis tool, sandboxed shells. |
Once an agent can read and write files, half of knowledge work becomes automatable: summarize this folder of PDFs, rename these photos by date, collect all invoices from 2025, refactor this codebase. Filesystem is the single tool that unlocks the most 'I wish I didn't have to do this' tasks.
Browser tools let agents use any web app. No API required. The tradeoff: clicking pixels is slow, brittle, and vulnerable to UI changes. OSWorld scores jumped from under 15% in 2024 to 72.5% by early 2026 after Anthropic's Vercept acquisition improved Claude's vision-based perception. Still, browsers remain the hardest domain to make robust.
When an agent needs to do real math, parse a CSV, transform data, or generate a plot, it writes code and runs it. ChatGPT's Code Interpreter, Claude's Analysis tool, and Devin's sandboxed environment all work this way. Much more reliable than asking the model to 'calculate in its head'.
MCP (Model Context Protocol) is the open standard launched by Anthropic and now backed by OpenAI and Google. It lets any AI client connect to any tool server with one protocol. As of April 2026, there are over 1,200 community MCP servers — think of them as apps for your agent.
Next we'll look at where those tools live — on someone else's cloud or on your own machine — and why the difference matters a lot.
15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-agentic-tools-overview-builders
What is the core idea behind "Tools an Agent Might Have: Filesystem, Browser, Code"?
Which term best describes a foundational idea in "Tools an Agent Might Have: Filesystem, Browser, Code"?
A learner studying Tools an Agent Might Have: Filesystem, Browser, Code would need to understand which concept?
Which of these is directly relevant to Tools an Agent Might Have: Filesystem, Browser, Code?
Which of the following is a key point about Tools an Agent Might Have: Filesystem, Browser, Code?
Which of these does NOT belong in a discussion of Tools an Agent Might Have: Filesystem, Browser, Code?
What is the key insight about "Captchas still win" in the context of Tools an Agent Might Have: Filesystem, Browser, Code?
What is the key insight about "MCP changed the game" in the context of Tools an Agent Might Have: Filesystem, Browser, Code?
What is the key warning about "Define the guardrails first" in the context of Tools an Agent Might Have: Filesystem, Browser, Code?
Which statement accurately describes an aspect of Tools an Agent Might Have: Filesystem, Browser, Code?
What does working with Tools an Agent Might Have: Filesystem, Browser, Code typically involve?
Which of the following is true about Tools an Agent Might Have: Filesystem, Browser, Code?
Which best describes the scope of "Tools an Agent Might Have: Filesystem, Browser, Code"?
Which section heading best belongs in a lesson about Tools an Agent Might Have: Filesystem, Browser, Code?
Which section heading best belongs in a lesson about Tools an Agent Might Have: Filesystem, Browser, Code?