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 · Payments & trading

com.getaxionlabs/axion-gateway

Verified answers with named sources: UK parking appeals, NYC dismissal rates, UK MTD tax facts.

Checks passed
11 / 22

3 failed · 8 not measured

Median response
20574 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
11

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Outdated check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?
410
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?
222

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

Inspect individual checks 22 recorded

Core

  • Fail
    Reachable

    Latest probe outcome is ok, HTTP 200.

  • Pass
    Valid handshake

    protocolVersion 2025-11-25, serverInfo.name axion-gateway.

  • Pass
    Lists tools

    tools/list returned 11 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 20574 ms over 1 samples in 7 days. The bar is 1000 ms. Need at least 5 samples.

  • Not measured
    Consistent (p95)

    95th percentile 20574 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

    29 of 29 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.

  • Fail
    Website matches publisher

    The website host does not match 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 2026-08-27T21:44:15.056373Z, within 180 days.

  • Not measured
    Open-source licence

    No GitHub repository to ask about a licence.

Start using com.getaxionlabs/axion-gateway

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 com-getaxionlabs-axion-gateway https://mcp.getaxionlabs.com/mcp

Capabilities

Tools 11

borough_odds

NYC hearing outcomes by borough/county: the share of decided hearings dismissed outright and dismissed-or-reduced in each borough, from the City's own records. Same ticket, different borough — measurably different odds.

Discovered
denial_clock

One call for a denied US health-insurance claim: both free appeal clocks with your actual dates where they can be computed, which external-review route applies (state, federal, or the federal one regardless of your state because the plan is self-funded), what the CMS file records for external reviews filed and overturned in that state and for that insurer — or 'not reported' where the file does not carry it — the insurer's own published appeal procedure where we have read it, and your state insurance department's consumer line with the date we read it from their own site. Use it at the second no as well, when the plan has said no again on internal appeal: external review is the step after it, and the CMS figures show how often that step overturned the insurer. Free, no key, attribution required. Source: CMS Transparency in Coverage Public Use File (PY2026 release), 45 CFR 147.136.

Discovered
get_mtd_answer

One MTD answer in full: the question, the short answer, the complete plain-text body, the FAQ pairs, and every GOV.UK/HMRC source page the answer rests on. Use the slug from search_mtd_answers.

Discovered
list_doors

The Axion Doors registry: every machine-callable door this estate operates, each with the trust block an agent should read before believing a tool result — the named source the data comes from, the validation method actually applied to it, the date the source was last checked, and an honesty grade saying where the measurement is weak. Also lists the free endpoints, the MCP servers, and the priced doors that are dormant (they answer HTTP 402 with an empty accepts[] until a payment rail is live). Filter by zone, or ask for one field across every zone.

Discovered
mtd_scope

One call for UK Making Tax Digital for Income Tax scope: give gross qualifying income (self-employment plus UK and foreign property, before expenses) and get whether you are in, from which date, which threshold and which tax return decided it, your first quarterly update deadline and all four dates, how many quarterly updates you owe (a foreign property is a SEPARATE business from a UK one), and the automatic exemptions. Answers the £20,000-or-less automatic exemption and the not-yet-but-soon case explicitly rather than returning nothing. Free, no key, attribution required. Every fact is traced to a named GOV.UK/HMRC page on mtdfacts.co.uk; the governing law is the Income Tax (Digital Obligations) Regulations 2026 (SI 2026/336).

Discovered
nyc_ticket

One call for a New York plate: every OPEN parking and camera violation live from the City's own Open Data API, what is owed and what is already in judgment, the citywide dismissal and dismissed-or-reduced rate for each ticket's violation type from the City's published hearing records, and the clocks — 30 days from the date on the ticket to request a hearing before late penalties start, roughly 100 days from issue before an unanswered ticket goes into judgment, and the $350 judgment-debt level at which a car can be booted. Disputing is free at nyc.gov and always has been. PRIVACY: the plate goes to the City's API and back inside this one request and is never stored or logged here. Free, no key, attribution required.

Discovered
parking_route

One call for a UK private parking charge notice: which independent appeals body hears this operator's appeals (POPLA for BPA members, the IAS for IPC members) with the published source for the window, all three deadlines computed from your dates (the discount window, the appeal to the operator, the independent appeal), and POPLA's own record for that operator — appeals decided, appeals allowed, the allowed share, the three-year series and how many the operator did not contest. Sending an appeal to the wrong body wastes the one free independent appeal and neither service forwards to the other, so an ambiguous operator name returns the candidates rather than a guess. Use it at the second no as well, when the operator has rejected the appeal and POPLA has refused it too: the record shows how that operator's decided appeals went, and a POPLA decision binds the operator, not the driver. Free, no key, attribution required. Source: POPLA Annual Report Appendices 2023-2025 and the published BPA and IPC member lists.

Discovered
popla_outcomes

POPLA's own record of how private parking appeals were decided, per operator, across its 2023, 2024 and 2025 annual reports. Give an operator for its three-year series and its change in percentage points; omit it for the all-operator totals and a ranking. These are decided appeals only — appeals an operator withdraws end in the driver's favour without a ruling and are counted separately.

Discovered
screening_dispute

Given a US tenant- or employment-screening company (Checkr, HireRight, First Advantage, Sterling, LexisNexis, ChexSystems, SafeRent, RentGrow, TransUnion Rental, AppFolio, RealPage, CoreLogic, Experian RentBureau, Early Warning), return that company's verified dispute route (portal URL, phone, email, postal address, each with the date it was read from the company's own site), the FCRA clock that applies after an adverse action (60 / 30 / 45 days, cited to statute), what the federal CFPB complaint record shows happened to people who complained about that company, and whether the screening company or the platform is the party that can actually fix the error. Free, no key, attribution required. Source: CFPB Consumer Complaint Database; narrative-derived counts cover 2011 - 14 Aug 2026, when the CFPB ceased publishing narratives.

Discovered
search_mtd_answers

Search 35 plain-English answers about UK Making Tax Digital for Income Tax — thresholds and start dates (£50k Apr 2026 / £30k Apr 2027 / £20k Apr 2028), quarterly update deadlines, digital records, penalties, exemptions, sign-up, software. Every answer is sourced line-by-line to named GOV.UK/HMRC pages, re-checked 7 October 2026. Returns the matching questions with short answers; call get_mtd_answer for the full text and sources.

Discovered
violation_dismissal_rates

How New York City parking and camera tickets are actually decided at hearing, violation type by violation type — the share dismissed outright and the share dismissed or reduced, computed from the City's own published hearing records (119.8M rows). Give a violation to get its rates (partial names match, e.g. 'SIDEWALK' or 'MUNI METER'); omit it for a ranking of all 80 violation types.

Discovered

Recent measurements

Probe history

View status
Latest stored probes. Raw probe rows are kept for 14 days.
CheckedResultResponseTools
Connected20574 ms11