An open-source AI research assistant for social scientists based on OpenClaw. Follow the build, the experiments, and every lesson learned along the way.
Fresh dispatches from the lobster lab
Telegram is the default. Everyone recommends it. Every tutorial starts with it. And that’s fine if you haven’t thought very hard about what you actually…
First attempt to use OpenClaw with a locally run model (Minimax M2.5) was successful. It’s not my claw, Bruce, just yet. My team’s young colleague…
Why we are building Academic Lobster in the open
Not everyone has the budget for a dedicated AI research assistant. We’re building one: open source, for social scientists, by social scientists. This blog documents every step: the setup, the models, the tweaks, the failures, and the wins. Everything shared on GitHub.
It is also important to openly discuss AI capabilities and how it can and inevitably will transform academia.
— Academic Lobster manifesto (a living document)
OpenClaw is an open-source platform that transforms AI from a passive text generator into an active, autonomous research assistant running on your machine. By granting the AI controlled access to your files, browser, computer, (email, cloud), it can independently execute multi-step background workflows such as scraping web data, formatting datasets, or running analysis scripts, based entirely on your high-level instructions.
github.com/openclaw/openclaw