awesome-open-science-software lists resources about open science and software:
- software as main experimental tool (aka computational methods)
- software as main study object (software engineering, programming language, systems software, ...).
- software support for open science (communication, data sharing, etc)
RSS feed: https://github.com/ASSERT-KTH/awesome-open-science-software/commits/master.atom
Author: Martin Monperrus & contributors
- Eclipse Science: open source and collaboration for computational science.
- Mozilla Science is a community of researchers, developers, and librarians making research open and accessible.
- Research Data Alliance (RDA) is a community-driven organization supported by the European Commission, the United States Government's National Science Foundation and National Institute of Standards and Technology, and the Australian Government’s Department of Innovation with the goal of building the social and technical infrastructure to enable open sharing of data.
- The Software Sustainability Institute's mission is to cultivate better, more sustainable, research software to enable world-class research.
- Openscience.org supports the development of open scientific software, in particular for cheminformatics.
- NumFOCUS Open-code, better science. Non-profit organization that promotes sustainable high-level programming languages, open code development, and reproducible scientific research.
Reference documents (edit to add one):
Events (edit to add one):
- IEEE Workshop on The Future of Research Curation and Research Reproducibility (Nov 2016)
- Dagstuhl Workshop: Rethinking Experimental Methods in Computing (March 2016) (organized by D. Delling, C. Demetrescu, D. Johnson, J. Vitek)
- Dagstuhl Workshop: Artifact Evaluation for Publications (Nov 2015) (organized by B. Childers, G. Fursin, S. Krishnamurthi, A. Zeller).
- Dagstuhl Workshop: Engineering Academic Software Dagstuhl Workshop (June 2016) (organized by C. Goble, J. Howison, C. Kirchner, O. Nierstrasz).
Artifact review and evaluations (edit to add one):
- Artifact evaluation (artifact-eval.org) (maintained by Shriram Krishnamurthi) discusses availability of experimental data and reproducibility of experiments.
- Artifact Evaluation at ISSTA'16.
- The Reproducible Science Project (reproduciblescience.org) is a joint-project by New York University, the University of California, Berkeley and the University of Washington to promote open-science.
- The ReScience journal is a peer-reviewed journal that targets computational research and encourages the explicit replication of already published research, promoting new and open-source implementations in order to ensure that the original research is reproducible.
Papers (edit to add one):
- Enhancing reproducibility for computational methods (Science, Dec 2016, Vol. 354)
- Journal of Open Source Software - JOSS The Journal of Open Source Software (JOSS) is a journal for research software packages. JOSS papers enable citation of research software packages with a DOI.
- Streamlining the Inclusion of Computer Experiments In a Research Paper, IEEE Computer, vol 55. no. 11, November 2018.
Tools (edit to add one):
- AgentFEM is an open-source finite-element workflow for computational experiments: editable Python models, pre-run checks, fields and time histories, with worked examples checked against analytical solutions. Runs locally on FEniCSx/PETSc/MPI; Apache-2.0 Python package. (documentation, teaching examples)
- Depsy helps build the software-intensive science of the future by promoting credit for software as a fundamental building block of science.
- CodeMeta The CodeMeta project proposes a way to have structured metadata for research software.
- code-ini Code meta data via .ini files
- MetaReview Free, open-source online platform for statistical meta-analysis in medical research. Features forest plots, funnel plots, GRADE assessment, and automated report generation. (GitHub)
- BGPT MCP Remote MCP server for searching scientific papers with structured experimental data from full-text studies. SSE + Streamable HTTP endpoints. (
npx bgpt-mcp) - Lune Research Remote MCP server grounding coding agents in full-text peer-reviewed papers from top venues (NeurIPS, ICLR, ACL, CVPR, USENIX Security), with citation trails in both directions and claim verification that returns the verbatim supporting quote. Streamable HTTP + stdio. (
npx @retrograde-labs/lune-mcp-server) - PRML / falsify Pre-Registered ML Manifest: an open specification (CC BY 4.0) for committing a machine learning evaluation claim — metric, comparator, threshold, dataset hash, seed, producer identity — to a SHA-256 hash before the experiment runs. Any retroactive edit changes the hash. Four byte-equivalent reference implementations in Python, JavaScript, Go, and Rust across 20 conformance vectors. JSON Schema in the SchemaStore catalog. Zenodo DOI 10.5281/zenodo.20177839; JOSS paper in preparation.
- Peptide Calculations Tools for peptide reconstitution, concentration, dilution and molarity calculations, with a browser calculator and reproducible Python and R implementations. (web calculator | documentation)
- UnitConv is a free browser-based unit converter for reproducible engineering and scientific calculations, backed by open conversion test vectors and rounding-validation tools. (QA dataset | rounding lab)
- K-Veritas Cryptographic verification for computational experiments: wraps any command and seals its results, code, hardware and timing in a signed report anyone can verify, without re-running the code (docs, releases, source).
- SciProof records computational experiments and creates machine-verifiable evidence manifests: integrity hashes (SHA-256), opt-in re-execution, numerical tolerances (rtol/atol), and direct LaTeX macro exports for paper drafts. Zero runtime dependencies, offline-first, Python 3.11+. MIT licensed (documentation, Zenodo DOI).
- celsus is an open-source Python workflow under AGPL-3.0. It reruns an arXiv paper's analysis on public data in a sandbox and reports whether the published numbers hold up.
Software data curation and preservation (edit to add one):
- Software Heritage project collects and preserves software. The project motivation is that "software embodies our technical and scientific knowledge and humanity cannot afford the risk of losing it." (project lead by Roberto Di Cosmo).
- tera-PROMISE - Research dataset repository specializing in software engineering research datasets.
- Zenodo - operated by CERN, contains several collections about software data:
- Data Archiving and Networked Services - DANS (The Netherlands)
Others:
- Awesome-msr: curated repository of software engineering data sets
- The Evaluate Collaboratory (managed by Matthias Hauswirth) is a hub for everybody interested in understanding and improving the state of practice in experimental evaluation in software science.
- The Github group "opensciences" (initiated by Tim Menzies) contains the source of http://openscience.us/, which is an "open science for software engineering portal".
- SoftwareEngineeringToolDemos This Github group gathers software prototypes associated with software-engineering publications (maintained by Emerson Murphy-Hill and his students).
- IEEE DataPort is an open repository of datasets hosted by IEEE
- Dataverse A data repository framework to share and publish research data
- Code as a Research Object is a prototype integration between Github and figshare.
- dokieli is a client-side editor for decentralised article publishing, annotations and social interactions for science. Its source code is open source under the Apache License, Version 2.0. Maintained by Sarven Capadisli (Github) et al.
- Scroll Press is a free, open-source preprint archive for web-native HTML research. Authors upload HTML and receive a permanent URL and a DOI. Every paper has a named human author intellectually responsible for the claims. Holds itself to ARCH 1.0, a public charter for what a web-native research manuscript must do. Part of The Aris Program.
- Open Science Framework (OSF.io) provides project management support for researchers across the entire research lifecycle. Developed by the Center for Open Science.
- Anonymous Github is a system to anonymize open-science Github repositories before referring to them in a double-blind paper submission.
- recipy records provenance for Python programs
- shournal records provenance on the shell. It provides the exact command which created a given file
- codeocean Code Ocean is a cloud-based computational reproducibility platform to run scientific code
- LabPal is a Java library that allows you to design, control, process and package experiments that are run on a computer, and to streamline the integration of results within a research paper.
- OpenChrom is a cross-platform chromatography data analysis tool with support for both proprietary vendor formats and open alternatives to publish as FAIR data.
- Calkit is a project framework and toolkit to help researchers work reproducibly.
- CiteMe is a free academic citation generator with a built-in reference checker that flags fabricated or hallucinated references. It searches 11+ databases (OpenAlex, PubMed, Semantic Scholar, CrossRef) and formats references in 40+ citation styles (APA, ABNT, MLA, Chicago).
- ValiChord — Distributed infrastructure for computational reproducibility validation built on Holochain.
- CiteVahti checks whether each manuscript claim is supported by the source it cites. Local-first (no manuscript upload), Zotero-integrated, human-first with AI assist.
- solveathome is an open platform where people set research directions on open problems, their own AI agents do the research, other contributors' agents check the results, and trusted human reviewers decide what is accepted, one vote per person. Every result, review and transcript is public, with a daily dump under CC BY 4.0. MIT licensed (GitHub).
- release-doi is a Claude Code skill that runs the release of a DOI-registered research repository on GitHub and Zenodo: it syncs the citation files (CITATION.cff, .zenodo.json, CHANGELOG) before the tagged release that mints the new version DOI, then writes that DOI into the citation fields and records the Software Heritage snapshot ID (SWHID). The skill text is in Japanese. MIT licensed.
- Evidence synthesis automation map maps 26 steps of a systematic review to the tools that already do them (link and price where verifiable, checked 2026-09), and reports a measured probe of what a web-enabled LLM did on nine real review chores, with scoring rules and limits stated. English and Chinese, CC BY 4.0.
- veripaper checks whether the references in a research draft actually exist (Crossref, DataCite, arXiv, Semantic Scholar, Europe PMC, DBLP) and labels each claim verified / cited / unverified / stale rather than guessing. Runs in the browser with no manuscript upload (live demo) or as a Python CLI, and ships a public citation-checking benchmark. MIT licensed.
- CRediT Matrix drafts CRediT contribution statements: assign the 14 roles in a grid, compare co-authors' answers with your draft, and export the statement, JATS4R XML or a contribution heatmap. Runs in the browser and works offline. MIT licensed (GitHub).
- Space Mushroom is a free open materials database (metals, polymers, ceramics, composites, space materials) in which each value is shown with its unit, test conditions, method and source. Data is CC BY 4.0, with CSV downloads, a dataset repository and a read-only remote MCP server (repo).
- Swedish PhD course "Tools for reproducible research". It gives very good pointers.
- Coursera MOOC on Reproducible Research
- REFUTE — Apache-2.0 benchmark for scientific critique & epistemic calibration on recent (2025–2026) science summaries. Separates critique skill from calibrated truthfulness (falsification, limitations, overclaims, missing-evidence refusal, confidence calibration, planted-flaw detection). Leaderboard · Technical report · Integrators