ctan.org with an AI agent
The short answer: ctan.org is the Comprehensive TeX Archive Network, where LaTeX packages live. We gave Claude Code a real task there, each in a brand-new session. With Playwright MCP a task used 167K tokens on average. With Dejo it used 127K the first time and 82K on every run after, 51% fewer, because Dejo replays a learned task without the model reading the pages again.
The task
- “search CTAN for the LaTeX class for writing knitting patterns (the package is called knittingpattern) and tell me its version”
What each tool used
| task | Dejo, first run | Dejo, every run after | Playwright MCP | Playwright CLI |
|---|---|---|---|---|
| 1 | 127K tokens, 18 s | 82K tokens, 1 call, 9 s, right | 167K tokens, 2 calls, 17 s, right | 214K tokens, 3 calls, right |
Calls are browser tool calls the agent made. Right or wrong is checked against the answer read off the live site before any tool ran.
Learned and replayed with Dejo on 13 November 2026
We ran the task again with Dejo on 13 November 2026, so you can see what it saves today. These are the steps Dejo kept, exactly as it reported them. Each one carries a check that it landed, and a replay walks them in plain code.
1. go https://ctan.org/
2. fill ref "e65" (by the words beside it)
3. press Enter
4. click link " Package knittingpattern" (by name)
5. answer (by position)Replay: 5 steps in 1.1 s, with no page reading. Answer: “2009-11-13” (right).
How we measured
Claude Sonnet 5 at medium effort, a brand-new Claude Code session per task and tool, @playwright/mcp 0.0.80, October 2026. The full method, and the sites that blocked every tool, are on the benchmark page. Raw data: benchmark-real-2026-10-21.json, benchmark-real-2026-10-23.json, benchmark-real-2026-10-23-fixes.json, benchmark-site-memory-2026-11-10.json.
Other websites we measured
- scrapethissite.com: a sandbox site for practising web scraping, with tables, forms and paging
- packages.debian.org: Debian's package directory, searched by name and release
- rfc-editor.org: the official home of the RFC series of internet standards
- catalog.ccv.edu: the course catalog of the Community College of Vermont
- practicesoftwaretesting.com: a demo tool shop built for practising test automation
- demowebshop.tricentis.com: Tricentis's demo web shop for test automation
- collection.internationalfolkart.org: the online collection of the Museum of International Folk Art in Santa Fe
- gutenberg.org: Project Gutenberg, a library of free ebooks
- pub.dev: the package site for Dart and Flutter
- hackage.haskell.org: the Haskell community's package archive
FAQ
How many tokens does an AI agent use on ctan.org?
In our test, Claude Code with Playwright MCP used 167K tokens per task on ctan.org on average. Dejo used 127K the first time and 82K on every run after, 51% fewer than Playwright MCP once the task was learned.
Can Claude Code automate ctan.org?
Yes. In fresh Claude Code sessions on one task there, Dejo answered 1 of 1 right, Playwright MCP 1 and Playwright CLI 1. With Dejo, a task done once is replayed later without the model reading the pages again.