factcircuit
FactCircuit: an auditable claim-and-evidence verification loop with immutable provenance, staged checks, bounded retrieval, and reproducible evaluation.
Chromascope: AI-assisted LC-MS analysis, across instrument vendors, with traceable results. Free and open source.
DataRust · PythonGitHub @chromascope-msOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
Open-source, AI-ready LC–MS workbench built in Rust. Interactive visualization, peak integration, batch quantification, and MCP-powered automation.
Taken from the repository README
Loading the full README.
Read the full README ↗Chromascope is a free, lightweight, easy-to-use, open-source, vendor-agnostic LC-MS workbench designed for targeted analysis. It opens mzML directly and works with any vendor file by converting it automatically with msconvert: download msconvert once and link it in Chromascope (how). Inspect batches, integrate peaks, build calibration curves, quantify and run QC. Use it like a normal desktop app, script it from the command line, or hand the controls to an AI assistant over MCP. The assistant does the reasoning; Chromascope does the maths, keeps the evidence and records exactly what happened.
https://github.com/user-attachments/assets/cfbe286f-962d-41b3-ad6f-ffa3f33bebf3
Video not playing? Open assets/demo.mp4 directly. It shows an AI client opening a file, extracting an XIC and previewing an integration in the Chromascope window; nothing is changed permanently.
test_file/..msi or portable .zip (Windows), .dmg (macOS), .deb, .rpm, .AppImage or .tar.gz (Linux).Details for every system, the checksums and building from source are on the download page. The desktop program checks GitHub once a day for a newer version; see Settings to turn that off.
| Program | What it is |
|---|---|
chromascope | The desktop app |
chromascope-cli | The same analysis engine as JSON on the command line |
chromascope-mcp | The desktop app as an MCP server: an AI assistant works in a visible window (74 tools) |
chromascope-engine-mcp | A window-less MCP server with the 14 calculation tools, for scripts and servers |
One engine sits behind all four, so a number from the window is the number from the CLI and from the assistant. Raw files are never modified, and results keep their units, settings, software versions and source hashes.
An assistant works in the window that its client started (chromascope-mcp). Its status bar shows how many requests the assistant made, with a link to the call log. When you ask it to propose instead of apply a change, for example new peak boundaries, the proposal appears in the window under Audit package, Review AI proposals, with its reason and evidence; only your Approve changes a result. Proposals saved in a project can be reviewed later in the plain chromascope.exe. See the guide to reviewing AI proposals.
cargo fmt --all -- --check
cargo clippy --locked --all-targets -- -D warnings
cargo clippy --locked --all-targets --all-features -- -D warnings
cargo test --locked
cargo test --locked --features mcp
cargo test --locked --no-default-features --features mcp-headless
cargo test --locked --all-features
The package needs Rust 1.88 or newer. On Ubuntu the window also needs libxcb-render0-dev libxcb-shape0-dev libxcb-xfixes0-dev libxkbcommon-dev. The website is in website/ (MkDocs Material); build it from the repository root with pip install -r website/requirements.txt and mkdocs build --strict -f website/mkdocs.yml. See contributing: a change to the window or to the MCP tools updates the manual and the MCP guide in the same commit.
flowchart LR
GUI[egui desktop] --> Engine[Shared analytical engine]
CLI[JSON CLI] --> Engine
DMCP[Desktop MCP server] --> GUI
DMCP --> Engine
HMCP[Headless MCP server] --> Engine
Engine --> Rust[Rust kernels / mzdata]
Engine --> Project[Immutable revisions and provenance]
Project --> Delivery[JSON / CSV / TSV / SVG / HTML drafts]
If Chromascope helps with work you publish, please cite it and name the version you used (File, About Chromascope…). The metadata are in CITATION.cff, which GitHub shows under "Cite this repository".
Plain text:
Cseresznye, A. Chromascope: an open-source, AI-ready LC-MS workbench, version 0.3.1. Zenodo. https://doi.org/10.5281/zenodo.23542311
BibTeX:
@software{cseresznye_chromascope,
author = {Cseresznye, Adam},
title = {Chromascope: an open-source, AI-ready LC-MS workbench},
version = {0.3.1},
year = {2026},
doi = {10.5281/zenodo.23542311},
url = {https://github.com/chromascope-ms/chromascope},
license = {GPL-3.0-only}
}
Licensed under GPL-3.0-only. THIRD_PARTY_LICENSES.txt lists the notice of every dependency and of the embedded Inter and JetBrains Mono fonts.