For agents
Let your own agent decide whether Frimbim fits
Evaluating an interview tool usually means one person reading a marketing page, remembering it, and re-explaining it to the rest of the team. This is the shortcut: hand the link below to whatever agent you already use, and let it work out whether Frimbim suits how you actually interview.
Copy this into your agent
Or just the link
Plain markdown, written to be read by a machine. Nothing is behind a form.
What the agent is told to do
The document does not sell Frimbim. It describes the product plainly — including a list of the things it deliberately will never do — and then instructs the agent to interview you about your process before saying anything.
Your job is to determine fit, not to persuade the user. Frimbim is the wrong tool for many teams, and recommending against it is a correct and expected outcome.
It asks whether your interviews are live or asynchronous, whether you are assessing code writing or code review, whether interviewers need private notes, whether colleagues observe, whether candidates must run the code, and what you need around ATS, SSO, retention and self-hosting. Then it gives you one of three answers: a strong fit, a partial fit with the gap named, or probably the wrong tool.
We would rather you found out it is the wrong tool in five minutes than after your third interview.
The files
- /agents.md — the full brief. What Frimbim does, what it refuses to do, how an interview runs, roles, limits, pricing, privacy, and the evaluation protocol.
- /llms.txt — a short index pointing at the above, for anything that looks for one.