| Version | Supported |
|---|---|
| latest | Yes |
| older | No |
Only the latest published release receives security updates.
Do not open a public GitHub issue for security vulnerabilities.
Report privately via GitHub Security Advisories.
Include:
- A description of the vulnerability and its potential impact.
- Steps to reproduce or a proof of concept (a small input file is ideal).
- The affected crate (
openproteo-io,openproteo-io-cli, oropenproteo-io-py). - The OS, Rust toolchain, and crate version you were running.
Expect an initial acknowledgment within 7 days.
In scope:
- Parser correctness on malicious input.
openproteo-iodispatches to vendor parsers and runs the canonical mzML writer fromopenproteo-core. Crashes (panics, OOB reads, infinite loops), arbitrary file writes, or memory corruption triggered by a crafted vendor file or mzML input are in scope. - Path-traversal or arbitrary-file-write bugs in
vendor2mzml(CLI) andopenproteo-io-py(Python wheel). - Supply-chain integrity of published artifacts on crates.io and PyPI: tampered manifests, missing provenance, unsigned releases.
Out of scope:
- Denial of service via legitimately oversized vendor files. Mass-spec acquisitions can be hundreds of GB by design.
- Vulnerabilities in third-party crates with no demonstrated exploit path through this stack. Forward those upstream.
- Issues that require write access to the parser source tree (this is a library, not a sandboxed service).
This repository is the umbrella of the OpenProteo stack. Reports about a specific vendor parser are usually better routed to the relevant repo - but the umbrella is a fine entry point and we will forward as needed:
- Thermo
.raw: OpenTFRaw - Bruker
.d/(timsTOF): OpenTimsTDF - Waters MassLynx
.raw/: OpenWRaw - Shared core: openproteo-core
We follow coordinated disclosure. Reporters are credited in the release notes unless they prefer to remain anonymous. We aim to ship a fix within 30 days of confirming a high or critical issue.
The vendor parsers in this stack were developed by clean-room reverse engineering of public-domain artifacts (PRIDE deposits, published specifications). They do not depend on any vendor SDK or binary blob. Bug reports about parser inaccuracy or unsupported acquisition modes are welcome but are not security issues - file those as regular GitHub issues on the relevant parser repo.