MCP.Market

Bestie Check MCP server

online · 100% uptime
by HireBestie·io.github.HireBestie/bestie-check·v0.3.0

Document reasoning checks: a helicopter view of claims and assumptions before handoff.

B81/100grade B
What users say
No reviews yet
Be the first
Safety scan
B81/100

full report

Adoption
Growing

0 stars6 downloads/wk

Reviews

Write one

Nobody has reviewed Bestie Check yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Bestie Check tools

Tool list not cached yet. `describe` through the gateway fetches it live.

Public scan report

scanner v0.1.10 · 2026-10-02 · same rubric, same numbers if you re-run it

1 medium
  • Code scan5 source files scanned20/25
  • Live reliabilityremote reachable in 7589ms (auth required)12/20
  • –Tool poisoningtools not inspected (endpoint requires auth); not countedn/a
  • Auth qualityOAuth resource metadata advertised on 40115/15
  • Maintenancelast push 0 days ago15/15
  • Maintainer identityregistry namespace matches repository owner7/10

Findings (1)

  • mediumeval / new Function usedexec.eval
    dist/cli.js: …ode, sch); const makeValidate = new Function(`${names_1.default.self}`, `${names_1.de…
Overall 81/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How Bestie Check compares

Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
81/100
higher than 62% of 48,255 servers
Uptime
100%
of our recent checks; the typical remote server answers 100%
Response time
401 ms
faster than 37% of 30,031 servers
Downloads a week
6
more than 1% of 12,191 servers with a package

Install Bestie Check in Claude Code, Cursor or VS Code

claude mcp add --transport http bestie-check https://www.hirebestie.com/api/check/mcp
Add to Cursor

What the publisher says

From the Bestie Check repository's README, as published. We do not edit it. Read it on GitHub

Bestie Check

See your document from above: every claim, and every assumption it rests on.

Bestie Check prepares a memo, report, or other document for a reasoning check. The free Quick Look gives you a helicopter view of its claims and assumptions. Agents can also read findings and a fix list from checks in the connected account. Paid checks require server availability and approval of a specific quote.

This repository contains the public plugin distribution. The Node CLI and stdio MCP implementation are bundled in @mybestie/check. Plugin version 1.2.0 uses CLI version 0.3.0, published on npm.

Connect

Hosted MCP with OAuth

For hosts that accept a remote MCP server, connect to:

https://www.hirebestie.com/api/check/mcp

Sign in to your Bestie account and authorize the connection. Use the hosted connection or the local stdio connection for each host. Configuring both duplicates the tools.

CLI and local MCP

Requires Node.js 20 or newer and npm.

npx -y @mybestie/check@0.3.0 login
npx -y @mybestie/check@0.3.0 capabilities
npx -y @mybestie/check@0.3.0 prepare --file memo.pdf --wait --progress
npx -y @mybestie/check@0.3.0 mcp

The plugin's local MCP process pins @mybestie/check@0.3.0. Sign in through the CLI first; credentials stay in the CLI's local credential store. You may instead configure BESTIEAPIKEY in your host's private environment. Never commit a key or include one in a prompt.

Claude Code

/plugin marketplace add HireBestie/bestie-check
/plugin install bestie-check@bestie

Use /bestie-check:verify to prepare a document. The optional Stop hook asks whether to offer Check; it never approves spending.

Cursor and Codex

The root plugin.json, mcp.json, and skills/ form a portable Agent Plugin. .codex-plugin/plugin.json provides Codex compatibility metadata. .cursor/mcp.json and .cursor/rules/bestie-check.mdc provide a Cursor configuration and rule for manual setup. Follow your client's local-plugin setup instructions.

Public directory acceptance is separate from this repository. This repository does not imply a listing in the Claude, OpenAI, or Cursor directories.

Agent skill

bunx skills add HireBestie/bestie-check --skill verify-deliverable

Expected journey

  1. Call inspect_capabilities and use the returned availability.
  2. Prepare the document with prepare_deliverable, or upload its bytes with the file-upload tools.
  3. Poll get_preparation until the returned nextAction is review; respect retry and failure states.
  4. Report the returned claims and assumptions without inventing findings or a verdict.
  5. For an existing check, read its findings and fix list using the account's authorized access.

Paid tools are offered only when the server reports them ready. An agent must obtain approval for the actual quote, its hash, and a credit ceiling. Installing a plugin, signing in, or saying "check this" does not approve spending.

Links

  • Connection guide
  • MCP documentation
  • CLI documentation
  • Support
  • Privacy

The fixture under review-fixtures/ is fictional and may be used for integration testing. Review-account credentials are supplied privately through submission portals; they are never part of this repository.

License

The plugin files in this repository are MIT licensed. Bestie's hosted service is governed by its own service terms. See LICENSE.

Nothing above is checked by us. What we check is on the safety report.

Bestie Check: common questions

Is Bestie Check MCP server safe?
Mostly: it is graded B (81/100). Read the Bestie Check safety report
How do I install Bestie Check?
It runs remotely at www.hirebestie.com. Add it to Claude Code, Cursor, VS Code or Claude Desktop with the snippets above, or call it through the mcp.market gateway without installing anything.
Does Bestie Check need an API key?
No key to paste: it signs you in with OAuth when your client connects.
Is Bestie Check maintained?
The last commit was 33 days ago (2026-10-02). The latest release is v0.3.0.
Is Bestie Check up?
100% of our last 28 checks got an answer. We check remote servers about four times a day.
How does Bestie Check compare with other MCP servers?
safety score 81/100 (higher than 62% of 48,255 servers); uptime 100% (of our recent checks; the typical remote server answers 100%); response time 401 ms (faster than 37% of 30,031 servers). Full comparison.