# ctan.org with an AI agent: 1 task measured

> A real task on ctan.org, each in a fresh Claude Code session: Playwright MCP used 167K tokens on average, Dejo 127K the first time and 82K on every run after. Steps, calls and right answers per tool.

Canonical: https://dejo.run/sites/ctan-org

> For AI agents: the index of this site is https://dejo.run/llms.txt, and every page is also served as markdown (add .md to its address, or send Accept: text/markdown).

ctan.org is the Comprehensive TeX Archive Network, where LaTeX packages live. Claude Code (Claude Sonnet 5) did this task there in brand-new sessions. Playwright MCP used 167K tokens per task on average; Dejo 127K the first time and 82K on every run after.

## Tasks

1. "search CTAN for the LaTeX class for writing knitting patterns (the package is called knittingpattern) and tell me its version"

## Tokens and calls per task

| task | Dejo, first run | Dejo, every run after | Playwright MCP | Playwright CLI |
|---|---|---|---|---|
| 1 | 127K | 82K, 1 call(s) | 167K, 2 calls | 214K, 3 calls |

## Learned and replayed with Dejo on 2026-11-13

Steps Dejo saved:

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, answer "2009-11-13".

## 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.

## Related

- [Dejo: browser automation MCP that remembers websites](https://dejo.run/)
- [Dejo MCP server setup: install in a minute](https://dejo.run/docs)
