
MachineWitness MCP server
Which MCP servers are MachineWitness's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.
read 2026-10-26
- Registry nameeu.machinewitness/archiveverified in the MCP registry
The registry name sits under machinewitness.eu. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.
- Stateanswerschecked 2026-10-26
- Tools6
- Server calls itselfmachinewitness
- Server version1.3.0
- Protocol version2025-06-18
On 2026-10-26 the server answered initialize and tools/list without a sign-in. First answer in 122 ms.
- Whether its tools do the product's job has not been judged yet.
- Checked on 1 of the 7 days it takes to judge reliability.
Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.
- INITE MCP scoreafter a weeka score needs 7 daily checks: 1 so far
- Answeredmeasuring1 of the 7 daily checks it takes to print a share
- Median first answer250 ms
- Space in an agent's context≈ 1,494 tokensthe definitions of all its tools (6), which an agent carries on every turn
- Tool definitionsA · 100 of 100
- Typed results0 of 6 toolsdeclared outputSchema: results come typed, read once
- Every tool passes all six checks.
What an agent can do with it
- not sure · retrieve past daily snapshots of a domain's AI opt-out files
- not sure · seal a dated copy of a web page before it changes
- not sure · check whether a domain's AI reservations are recorded each day
Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?
Worth a look
None of the five warning signs in these definitions.
A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →
check_coverageAnswer whether this archive observes a given domain, from which sealed day onward, which machine-readable files it requests, and at what cadence. Returns metadata only — never the contents of any file, and never an assessment of the…
suggest_domainSuggest that a domain be observed. This is a request, not an instruction: admission follows documented criteria (a connection to the European Union, publicly served machine-readable files) and is decided by the operator. The answer is…
order_captureCreate a payment link for a point-in-time record: both witnesses fetch the named addresses, take a full-page picture of each, and give each capture its own qualified electronic time-stamp. Up to twenty public addresses, all on the same…
order_extractCreate a payment link for an evidence extract: everything this archive recorded on a domain, as one verifiable document (every observed day, every file type, both witnesses, with the seals and verification instructions). One to twenty…
order_observationCreate a payment link for the daily observation of domains, as a yearly or monthly subscription: from admission, within five working days of payment, two independent witnesses record each domain every day in full (homepage and the…
capture_statusWhere a point-in-time record stands, by its reference (MW-K-…): awaiting payment, queued, capturing, first delivery sent, complete. Per witness: waiting, capturing, captured, failed or refused. Never returns the e-mail address, the…
How to connect
Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:
{
"mcpServers": {
"archive": {
"type": "http",
"url": "https://machinewitness.eu/mcp"
}
}
}Add it
Claude Code, in a terminal
claude mcp add --transport http machinewitness https://machinewitness.eu/mcpclaude.ai, ChatGPT and other clients: add https://machinewitness.eu/mcp as a custom connector in the settings.
Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.
Cookies
Before we count this visit
Analytics and ad measurement tell us which pages and which of our ads do their job: Google Analytics and Google Ads, the Meta pixel, Yandex Metrica, and one first-party id of our own. Nothing is stored until you answer, and declining changes nothing about the site.