Help wiki

A free, open-source command-line assistant for academic research. It lives inside your project folder: it harvests papers from publisher databases, organises your BibTeX references, and builds your documents (LaTeX, Word, PDF). It is not a web app — this website is only the signpost; the tool runs in your terminal.

Windows 10/11, macOS, or Linux, and five minutes. The one-line installer checks everything else itself and tells you if anything is missing.

With the one line for your platform on the install section of the home page. The installer tries GitHub first and falls back to the GitLab mirror automatically, so a blocked network is not a dead end.

Open a terminal inside your research folder and type clt. Everything claudito does happens in that folder — it never wanders elsewhere on your disk.

Inside the session, type new project. Claudito creates the structure — Project plan/, Documents/, references/, latex/ — and a guided tutorial walks you through the first steps, checking that each one is really done.

In Project plan/Project.txt — one plain-text master file. Markdown-style headings, a few LaTeX fragments if you want them, and lines starting with // for private notes that never appear in the output. You never write LaTeX unless you choose to.

Type build. Claudito compiles the master file into latex/build/main.pdf, a Word file, and the LaTeX source — all three from the same text, always in sync.

Type writer and a local writing window opens in your browser (still offline, still your machine). You write plain text, pick a journal or conference template from the list, and claudito applies the formatting — you never touch the journal's style rules.

Yes — type presentation and claudito builds a slide deck as PDF from the same master document. Same content, another format, no PowerPoint.

Type harvest with your topic, and claudito searches the publisher databases (Scopus, ScienceDirect, Springer, IEEE, ACM, Wiley…), downloads the papers into references/ organised by theme, and creates a clean BibTeX entry for each. The curated shortlist stays separate from the harvested pool.

Claudito checks for new releases and can update itself. You can always re-run the latest install line instead — your projects are never touched by an update.

Only literature-search queries go out, to the publisher databases you asked to search. Your text, your references and your builds stay on your machine. There is no account, no telemetry, no cloud.

Claudito archives Project.txt before every change — the Project plan/archive/ folder is a full history of snapshots. You can always go back; nothing is ever overwritten without a copy.

Yes — a claudito project is a plain folder. Version it, zip it, sync it, carry it on a USB stick. Claudito also has its own team sync if you prefer to keep everything inside the tool.

Yes. Claudito can be set up across a lab's internal machines, with controlled outputs to personal computers — either through the command line or through a dedicated interface built for your setup. This kind of deployment is done together with the owner, not self-serve: get in touch describing your lab and what you need.

In clauditoArchive/logs/ inside the project. Every session is logged there, including what claudito did and why. Most mysteries end in those files.

Open an issue on the claudito-site repository. Describe what you typed, what you expected, and what happened — a copy of the relevant lines from clauditoArchive/logs/ helps enormously.