MCP HeartbeatTHE MCP OBSERVATORY
Public workspace

Explore

Evidence for your next connection.

Independent measurements of remote MCP servers. Every observation has a timestamp.

Read-only probes

MCP server · Data access / search

Near Fine

Search booksellers worldwide for any book, new or used, and set alerts for copies not listed yet.

Checks passed
13 / 22

1 failed · 8 not measured

Median response
222 ms

Median of successful probes over 7 days

30-day uptime
Not measured

Share of countable probes over 30 days, including daily rollups. Auth challenges are excluded.

Discovered tools
5

Tool definitions returned by this server

Independent monitoring

Health & checks

Status history

Outdated check

Last connection measurement is over 24 hours old. Current availability is unknown.

Last checked
Check summary by test group
Test areaPassedFailedUnmeasured
ConnectionCan a client connect and discover tools?
500
Reliability & speedDoes it respond consistently and quickly?
004
Access & authenticationWhat is required to use this server?
102
Tool qualityAre the tool definitions usable by an agent?
400
Publisher identityCan the publisher be traced to its source?
312

Unmeasured means there is not enough evidence to decide. How these checks work

Inspect individual checks 22 recorded

Core

  • Pass
    Reachable

    Latest probe returned HTTP 200 in 222 ms.

  • Pass
    Valid handshake

    protocolVersion 2025-11-25, serverInfo.name nearfine.

  • Pass
    Lists tools

    tools/list returned 5 tools.

  • Pass
    Current protocol

    Negotiated 2025-11-25. The bar is 2025-06-18 or newer.

  • Pass
    HTTPS

    The endpoint URL is https.

Performance

  • Not measured
    Fast (p50)

    Median latency 222 ms over 1 samples in 7 days. The bar is 1000 ms. Need at least 5 samples.

  • Not measured
    Consistent (p95)

    95th percentile 222 ms over 1 samples in 7 days. The bar is 3000 ms. Need at least 20 samples.

  • Not measured
    Reliable (uptime)

    1 of 1 attempts returned a tool list in 30 days (100%). The bar is 99%. Need at least 30 samples.

  • Not measured
    Stable tool list

    Fewer than two successful probes in 30 days, so tool-list stability is not measured.

Access

  • Pass
    Open access

    tools/list succeeded with no credential, and the record declares none.

  • Not measured
    Standard auth

    Auth is not required, so the OAuth metadata check does not apply.

  • Not measured
    Free tier cited

    No publisher quote has been checked.

Tools

  • Pass
    Tools described

    Every tool description is at least 20 characters.

  • Pass
    Valid input schemas

    Every inputSchema is a JSON Schema object.

  • Pass
    Parameters documented

    6 of 6 parameters have a description (100%). The bar is 80%.

  • Pass
    Safety hints

    Every tool sets readOnlyHint or destructiveHint.

Publisher

  • Pass
    Verified namespace

    Listed in the Official MCP Registry.

  • Pass
    Website matches publisher

    The website host matches the reverse-domain namespace.

  • Fail
    Source repo linked

    No repository URL is on file.

  • Not measured
    Repo owner matches

    This namespace is not an io.github login, so the owner comparison does not apply.

  • Pass
    Recently updated

    Updated 2026-10-28T15:23:26.456206Z, within 180 days.

  • Not measured
    Open-source licence

    No GitHub repository to ask about a licence.

Start using Near Fine

Connect to this server

Choose your client, copy the configuration, then enable the server in your client’s MCP settings.

Run this command in your terminal, then restart Claude Code.

Claude Code

claude mcp add --transport http near-fine https://nearfine.com/mcp

Capabilities

Tools 5

cancel_alert

Cancel one of the person's alerts on Near Fine by its id, from list_alerts or set_alert, deleting it as its Delete button on Near Fine does. The first time, they sign in. Returns the cancelled alert and the person's plan.

Discovered
list_alerts

List the person's alerts on Near Fine. The first time, they sign in. Finds come first, then alerts being checked, then the rest, newest first. Each has its id, query, country, status, when it was set, last checked and runs until, a link to it on Near Fine, how many times it has emailed a find and when it last did, and, once it has found a listing, the cheapest it last found: its title, seller, condition, price and shipping in the person's currency, and the URL to buy it. Statuses: pending (noting what's listed already), live (watching, as a Pro alert stays after a find), found (emailed a find and stopped), ended (the person got the book or stopped it), expired (ran its six months) and paused (the plan has room for fewer alerts). Also returns the person's plan.

Discovered
list_sources

List the bookseller catalogs Near Fine searches, each with the countries it serves. Pass a country to see only the sources that serve buyers there.

Discovered
search_books

Search for a book across every bookseller Near Fine covers for the buyer's country, ebook and audiobook stores included; list_sources names them. Shops' ebooks and audiobook downloads are offered only in their own country. Pass an ISBN-10 or ISBN-13 for an exact edition, or title and author keywords. The query can also carry a price cap, as in "under $40" or "max 30 €" (in the buyer's currency when it names none), and ask for a first edition, a signed copy, a dust jacket, a hardcover or a paperback, as in "wolf hall signed first edition under £200", or for an ebook or audiobook alone. Offers over the cap, counting known shipping, and offers in other formats than the one asked for are left out; criteria reports what was read. Returns offers sorted cheapest first, unpublished shipping from abroad counting as the price again, each with title, seller, condition (new, used, ebook or audiobook), price and shipping in the seller's currency, the same in the buyer's currency when it differs, and the URL to buy. Shipping is left out where the shop doesn't publish it. Each offer may carry a match from 0 to 1: how sure a decision model is that the listing is the book searched for rather than a study guide, summary, adaptation or another book that shares a word. Offers below 0.2, which rarely are, are left out and counted in doubtful. When the query asks for attributes, each offer carries verdicts: true where the decision model, reading the listing and any filter the bookseller applied, confirms the listing claims it, false where it doesn't, and left out where unjudged. Also reports which sources answered, failed or timed out, a cover image of the best match, an amazon_url that opens the same search on the buyer's Amazon site, and a search_url that opens it on Near Fine, where the buyer can set an alert and get an email when a new listing turns up with everything the query asks for. A search takes a few seconds.

Discovered
set_alert

Set an alert on a search for the person. The first time, they sign in to Near Fine. Near Fine checks the alert for six months and emails them when a new listing turns up with everything the query asks for, such as a first edition, a signed copy or a price cap. On the Pro plan the alert keeps watching after a find and emails again when newer listings turn up; on other plans it stops after one email. The query reads as search_books reads it. Listings there already never count: the alert stays pending until its first check has noted them. Setting an alert that's already set keeps it as it is. Returns the alert, with its status and its link on Near Fine, and the person's plan: how many alerts it allows, how many are in use and how often each is checked. A full plan comes back as an error; paid plans, at plans_url, have more.

Discovered

Recent measurements

Probe history

View status
Latest stored probes. Raw probe rows are kept for 14 days.
CheckedResultResponseTools
Connected222 ms5