---
title: Scholar Sidekick for WebMCP
description: Scholar Sidekick registers seven read-only citation tools on navigator.modelContext, so an in-browser AI agent can verify, audit, resolve, format, and export citations by typed tool call instead of scraping the page. No install, no API key.
doc_version: "1.0"
last_updated: "2026-09-02"
---

# Scholar Sidekick for WebMCP

> Seven citation tools declared to the browser's own AI agent. No install, no account, no API key.
> Last updated: 2026-09-02
> HTML version: https://scholar-sidekick.com/integrations/webmcp
> Standard: https://developer.chrome.com/docs/ai/webmcp
> W3C community list: https://github.com/webmachinelearning/awesome-webmcp

An agent that reads a citation page has to interpret rendered text, and it can misread it. WebMCP removes that step. The page hands the agent a **typed tool it can call directly**, and the answer comes back as structured data from the same endpoints that power every other Scholar Sidekick integration.

> This is a transport, not a new product. The seven tools are the same ones the [MCP server](https://scholar-sidekick.com/mcp) offers to agents outside the browser. WebMCP removes the install step for an agent already in the tab.

**Experimental.** WebMCP is an emerging standard and the browser API is still changing.

## How it works

1. You open `scholar-sidekick.com` in a browser whose in-browser Model Context API is enabled.
2. An inline script feature-detects `navigator.modelContext`. If the API is absent it stops, and the page behaves exactly as before.
3. If the API is present, the script registers the seven tools below, each with its own JSON input schema.
4. The browser's agent calls a tool by name. The call reaches the public Scholar Sidekick API over HTTPS and returns structured data.

## The seven tools

| Tool | What it does |
|---|---|
| `verifyCitation` | Checks a claimed citation against the record its identifier actually resolves to - the real-identifier-plus-invented-title pattern documented by Topaz et al. (Lancet, 2026). |
| `auditBibliography` | Audits a whole bibliography (BibTeX, RIS, CSL-JSON, or a claims array) in one call: a verdict per entry plus a retraction check and a corpus summary. Capped at 25 entries. |
| `checkRetraction` | Reports whether one work is retracted, corrected, or carries an expression of concern (Crossref / Retraction Watch). |
| `checkOpenAccess` | Reports one work's open-access status and its best legal full-text URL (Unpaywall). |
| `resolveIdentifier` | Turns a DOI, PMID, PMCID, ISBN, arXiv ID, ISSN, ADS bibcode, or WHO IRIS URL into structured CSL-JSON metadata. |
| `formatCitation` | Formats identifiers into a finished citation: Vancouver, AMA, APA, IEEE, CSE, or any of 10,000+ CSL styles. |
| `exportCitation` | Exports identifiers to a bibliography file: BibTeX, RIS, EndNote, RefWorks, NBIB, Zotero RDF, CSL-JSON, or CSV. |

All seven are read-only. None of them writes data, takes payment, or changes anything you own.

Some third-party directories still list six tools. `auditBibliography` joined the surface after they indexed the site; the count here is current.

## What it does not do

- It does not search for papers. Give it an identifier or a claimed citation; it will not find new literature for you.
- It does not read your page, your tabs, or your files. Only the arguments the agent passes to a tool are sent.
- It does not run in browsers without the API - which today is nearly all of them. Those visitors see no change at all.

## Privacy

- Only the arguments of a tool call leave the browser - nothing else on the page.
- Requests go to `https://scholar-sidekick.com` over HTTPS.
- No account and no API key; the anonymous tier is used by default.
- Inputs are not retained or used for training. See the [privacy policy](https://scholar-sidekick.com/legal/privacy) for full disclosure.

## About the standard

WebMCP is described in the [Chrome developer documentation](https://developer.chrome.com/docs/ai/webmcp) and discussed in the W3C Web Machine Learning community group, which maintains a [list of sites that implement it](https://github.com/webmachinelearning/awesome-webmcp) - Scholar Sidekick among them. The standard is young, so treat this surface as experimental: the tools are stable, the way a browser exposes them is not.

## Related

- [Integrations index](https://scholar-sidekick.com/integrations) - all first-party integrations
- [MCP server](https://scholar-sidekick.com/mcp) - the same seven tools for agents outside the browser
- [Citation verifier](https://scholar-sidekick.com/tools/citation-verifier) - the same fabrication check as a free web tool
- [API docs](https://scholar-sidekick.com/docs) - the REST endpoints every tool calls

## Sitemap

See the full [sitemap](https://scholar-sidekick.com/sitemap.md) for all pages.