MachineWitness
Server Details
EU archive of the machine-readable web: check a domain, order a time-stamped record or extract.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- martinschenk/machinewitness-mcp
- GitHub Stars
- 0
- Server Listing
- MachineWitness
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 6 tools
Each tool has a distinct target: per-reference status, per-domain coverage lookup, and three order_* products with clearly different scopes (point-in-time record, historical extract, daily subscription). The three order_* tools share the identical 'create a payment link' framing and could be momentarily confused at a glance, but the descriptions spell out distinct inputs (addresses vs. domains vs. subscription) and refusal rules.
All six tools follow a strict verb_noun snake_case pattern: capture_status, check_coverage, order_capture, order_extract, order_observation, suggest_domain. No deviations in casing or style.
Six tools is well-scoped for a service offering three orderable products plus status, coverage and suggestion lookups. No filler tools and no obvious missing companion operation at this count.
The ordering and lookup surface is solid, but lifecycle gaps remain: there is no way to verify a delivered record or extract's seal/reference, no tool to list or retrieve one's own orders, and no operation to cancel the observation subscription even though cancellation semantics are described. Agents can work around these only via e-mail, which is a real dead end for programmatic verification.
Available Tools
6 toolscapture_statusStatus of a point-in-time recordARead-onlyIdempotentInspect
Where 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 addresses ordered or anything else about the customer.
Inputs: reference (required).
check_coverageCheck whether a domain is observedARead-onlyIdempotentInspect
Answer 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 domain or its operator. 'Not observed' is a statement about this archive, not about the domain: most of the web is not observed.
Inputs: domain (required).
order_captureOrder a point-in-time record of public addressesAInspect
Create 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 domain. Before the link is created, every address is checked at no charge: a public domain name, not on the stop list, robots.txt allows our crawler, and it answers without a login; refused addresses are named with the reason. This call charges nothing. The capture starts only after a person has paid on the returned page; the record is then delivered by e-mail, and a second e-mail follows after the night's seal. Only addresses are asked for, never a purpose. Fees: https://machinewitness.eu/services
Inputs: addresses (required), email, language, timezone.
order_extractOrder an evidence extract for observed domainsAInspect
Create 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 domains. A domain this archive never observed is refused by name before the link is created, because there would be nothing to deliver; use check_coverage first. This call charges nothing. After a person has paid on the returned page, a confirmation with the reference follows at once and the extract is normally delivered by e-mail within five working days. Only domains are asked for, never a purpose or a matter. Terms: https://machinewitness.eu/extract · Fees: https://machinewitness.eu/services
Inputs: domains (required), email, language.
order_observationSubscribe domains to daily observationAInspect
Create 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 machine-readable files), seal each day and anchor it externally; the admission is entered in the public record. The customer pays for the measurement, not for a document: the data in hand is an evidence extract (order_extract). A domain already observed daily in the core is refused by name, because it costs nothing. On cancellation the subscription ends with the paid period and everything witnessed until then stays sealed. This call charges nothing; the subscription starts only after a person has paid on the returned page. Fees and tiers: https://machinewitness.eu/services
Inputs: domains (required), email, interval, language.
suggest_domainSuggest a domain for observationAInspect
Suggest 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 always 'received' with 'guarantee: none' — it creates no obligation to observe the domain, no timeline, and no assurance that it will be added. A reason is required and is read by a person.
Inputs: email (required), domain (required), reason (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Wayback Machine snapshots, on-page SEO audits and Wikidata facts about companies and people
Common Crawl's public web archive — find every time a URL was crawled since 2008 and read the…
Wayback Machine for agents: closest archived copy of a URL with its text, and full capture history
Parsed WHOIS, RDAP, DNS, certificates, nameservers and domain history for any domain.
Related MCP Servers
- AlicenseAqualityDmaintenanceMCP server for the Internet Archive's Wayback Machine. Search archived snapshots, extract page text from a specific date, track how a site has changed over time, check if broken links are recoverable, and perform research across Internet Archive collections.63MIT
- AlicenseNot gradedqualityBmaintenanceEnables searching Common Crawl's web archive for crawl history of any URL since 2008. It also retrieves the exact captured page bytes from the archived records.110 npmMIT

siteglass-mcpofficial
AlicenseNot gradedqualityCmaintenanceEnables archival of any public URL as interactive snapshot (rrweb, PDF, PNG) and autonomous web QA for apps (register, scan, generate and run end-to-end flows).49 npmMIT- AlicenseNot gradedqualityCmaintenanceEnables agents to retrieve the closest archived snapshot of a URL with extracted page text, and to query the complete capture history for a URL, prefix, host, or domain, with fallback to Common Crawl.57 npm1MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.