How e2er works in practice
What stays on the researcher's computer
You can run, export and check research without an account. Publishing is optional and adds a public description; the files stay where you keep them. The AI models are hosted services, so the text sent to them leaves your machine. Every connector states what it sends out.
| Task | Command | Status | Local or not |
|---|---|---|---|
| Run a template | e2er run | In place | Workspace and run database stay on your machine. Prompts go to the AI provider you choose. |
| AI providers | Anthropic API, OpenRouter API, Claude Code, Codex, Gemini CLI (not tested) | In place | All five are hosted. Full runs are tested live on Claude Code, Codex and OpenRouter; Gemini needs a Gemini API key and is not tested. A provider that runs on your own computer is not supported yet. |
| Literature library | e2er library | In place | PDF files and the library database stay local. A passage is sent to the AI provider when a specialist reads it. |
| Export the results | e2er export | In place | Paper, results, code and a fingerprint of every file, written to a local folder. |
| Check the results | e2er verify | In place | Runs without an account or API key. |
| Check citations online | e2er verify --online | In place | Optional. Sends the reference list to OpenAlex, Crossref or Semantic Scholar. |
| Prepare publication | e2er publish | In place | Writes a description of the study to a local file, e2er.json. In e2er 0.10.0 and later. |
| Publish on e2er | e2er publish --to | In place | Optional. Adds the description to the public list; the files stay where you keep them. |
| Record AI usage | e2er publish | In place | Calls, models and tokens per specialist come from the run database and go into the publication description. The exported folder does not carry them. |
| Fingerprint the input data | e2er export | In place | The export holds the data the run acquired (data.db) with its fingerprint, the data dictionary and summary statistics. A study may leave the data out of what it shares; the Bitcoin study shares the dictionary and summary only. |
Governance
Changes to the formats are proposed and discussed on the proposals page. How a steering group is formed and when the formats move to a neutral body are still open.
Background: Brunsson and Jacobsson, A World of Standards, Oxford University Press, 2000; O'Mahony and Ferraro, "The emergence of governance in an open source community", Academy of Management Journal 50(5), 2007, 1079–1106.
Risks
| Risk | Current response |
|---|---|
| Checks favour work that a machine can check, such as numbers. | Each template lists which checks apply to its method; the others are marked not applicable. |
| Researchers hand over judgment to the specialists. | Researchers decide at which steps the process pauses for their review. A pause can be set after any step (e2er run --review-at); at a pause the researcher can approve, edit files, give instructions or send a step back (e2er review). |
| Dependence on a few AI providers. | Five providers are supported, and one question can be run on several and compared (compare runs). |
| Usage figures invite gaming. | The site shows no scores or rankings of people. Only published studies count in figures. |
| A study can pass every check and still be weak. | A study marked checked has numbers and citations that trace to their sources. Readers and reviewers judge its quality. |
Data protection
e2er does not upload your data. The datasets in Discover are public deposits on Zenodo and Harvard Dataverse. Approval and consent for research with personal data are not recorded yet; a field for them in the study description is planned.
Credit
Published studies list their authors with CRediT roles and cite the templates, specialists, skills and datasets they used. DOIs for public data and code are deposited on Zenodo with the researcher's own account (e2er publish --zenodo). A DOI for the study itself is planned.
A study published with e2er publish --demonstration carries a Demonstration label, shown as a notice on its page and its dossier and as a footnote on the paper's first page.
Formats
Record: a study produced without enforced checks
On 5 August 2026 e2er produced one study with its checks recording problems without stopping the process. Full record.