Examples/MS/MS de novo

MS/MS de novo

Created from a prompt

OpenCode (DeepSeek V4 Pro via OpenRouter) in a Vercel Sandbox. Starter only (AGENTS.md, serve.py, player/, helpers/dom/view_kit). No hardcoded demo code. Cursor did not edit workspace/.

2026-10-15 · task _2LBQYGPGWs6 · opencode · openrouter/deepseek/deepseek-v4-pro-0813

The prompt

Build a peptide sequencing exercise from a tandem mass spectrum. I am given a fragmented tryptic peptide: it ends in K or R. The spectrum is a set of peaks. Positions are trustworthy. Heights are only a rough guide, so I should sequence from mass gaps, not from which peak is tallest.

I can click two peaks and see the mass between them, and which amino acids that gap could be. I can zoom. I type a sequence in one-letter code and submit. I get a few tries. Leucine and isoleucine count as the same. Scoring should care about whether my sequence breaks in the same places as the real peptide, and a little about whether the total mass matches. An exact sequence, ignoring that leucine swap, should be a full score. After the last try, label the peaks and show the sequence. Several spectra. The same one should be the same peaks.

Live environment

Running the prompt-created workspace on this machine — not a mock. Save an episode, then open Replay below to scrub the timeline.

opening · 0 ms

Generated workspace

Exact files the agent wrote — not rewritten by hand.

# AI : shared Action contract — the page posts {"move": "..."}; keep this shape.
from pydantic import BaseModel


class Action(BaseModel):
    move: str