Voidly Hosted MCP

Look up the same durable operation ID without dispatching mail. Requires an existing inbox's agent key as the Bearer credential; this server cannot create an inbox.

{"destructiveHint":false,"idempotentHint":false,"openWorldHint":true,"readOnlyHint":false,"title":"Submit one plain-text message"}

{"destructiveHint":false,"idempotentHint":true,"openWorldHint":false,"readOnlyHint":true,"title":"Look up send status"}

{"destructiveHint":false,"idempotentHint":true,"openWorldHint":false,"readOnlyHint":true,"title":"Read sending limits"}

{"destructiveHint":false,"idempotentHint":true,"openWorldHint":false,"readOnlyHint":true,"title":"Check Voidmail inbox setup"}

Read rolling 24-hour rawPublic probe metrics and lifetime Voidly evidence counters as boundedstatistics JSONare pages.temporarily Startwithheld withoutpending a cursor; follow next_cursor for every by_domain row. Last-probe time differs from fetch time; community-path rows can beprivacy-reviewed Voidly-operatedaggregate andprojection. doThe nottool provereturns independentpublication_held operatorswithout orreading acceptedprobe measurements.rows.

How much of Voidly's measured-domain corpus carries a Citizen Lab content category (~69% via the 14k-domain Citizen Lab list) — the observatory disclosing its own categorization blind spot. Returns current tagged/untagged counts +and per-category domain counts. Use to gauge how complete ⟨5 unchanged words⟩ 

 ⟨19 unchanged words⟩  national blocks per region. HONEST: confirmed-block counts are a measurement-density map, not a censorship ranking (e.g. Africa showsdepend 0on confirmedmeasurement blocksdensity despiteand athe high>=3-network blockconfirmation fractiongate, because theso >=3-networkthey confirmationare gatenot needsa densecensorship coverage).ranking. Use block_fraction + countries_measured for context.

 ⟨5 unchanged words⟩  with rankings for monitored countries. Returns country scores derived from OONI anomaly rates, risk tiers, and measurement counts aggregated from OONI, IODA and CensoredPlanet overfor the recent measurementsource-declared window. Read current coverage and dataAsOf from the response.response; scores are not confirmed block counts.

 ⟨14 unchanged words⟩  with confirmed national blocks, the deepest censor (Iran,reported ~789by domains)the current response, the broadest-blocked category, measurement depth + data ⟨32 unchanged words⟩ 

How accurate is Voidly's v3.3 censorship classifier? TheRead honestthe answercurrent training sidecars under THREEthree evaluation regimes of increasing difficulty, in one call: stratified-random (in-distribution upper bound, AUC ~0.90 / F1 ~0.73), leave-country-out (cross-country generalization, F1 mean ~0.71 / median ~0.87 over 127 countries), and forward-temporal (train past / predict future, AUC ~0.67). /The F1response ~0.47)gives —AUC, plusF1, the generalization gap (delta AUC -0.23, 'DEGRADES forward') and which metric to cite for which use. Reads the live training sidecars. Use when asked 'how accurate is the model?' — neverNever quote one number alone. HONEST: the retired v2 '0.998 F1'evaluation had country-tier leakage and is not a live claim.

Country-day censorship classifier score (GradientBoosting v3.3,). Read the current LOCO mean F1 0.711and acrosssmall-sample countries;median thecaveat LOCOfrom median/v1/classifier/info; 0.870this is inflated bynot small-sampleshutdown-onset countries).skill.

 ⟨24 unchanged words⟩  (ANON/VPN), which strips coincidental gambling/adult overlap. Iran-Russia share 135 domains incl. 23 of the same human-rights orgs + 25 of theRead samecurrent VPNoverlap toolscounts (4.6xfrom the next pair on political).response. HONEST: overlap is correlation, NOT proof of coordination; China under-counted.

 ⟨44 unchanged words⟩ the confirmed-national layer (>=3 independent networks) — countscompare agreecurrent counts with get_national_blocklist + get_co_blocking (e.g. IR vs RU: IR 789 blocked, 135 shared, RU blocks more NEWS while IR blocks moreand human-rights/VPN).get_co_blocking. Use for 'how does country X's censorship differ ⟨33 unchanged words⟩ 

 ⟨20 unchanged words⟩ Pass country=XX&asn=NNNN for one network's per-category block rates +and a blanket/selective/permissive/mixed label (e.g. KZ AS207446 blocks 94% of VPN/circumvention tools but 0% of news — a 'VPN-blocker';label; RUpass AS47541country=XX isfor the opposite, blockingcountry's newsnetworks notranked. VPNs);Read passcurrent country=XXpercentages forfrom the country's networks ranked.response. HONEST: block_rate is over MEASURED domains per  ⟨26 unchanged words⟩ 

 ⟨27 unchanged words⟩ bounds what the >=3-network confirmed-block gate can confirm. RU 510 / IN 80 / IR 50 networks;Read ~42current countriescoverage carryfrom ASNthe tags.response. HONEST: coverage depth, is NOT a censorship score; an absent/shallow country is ⟨8 unchanged words⟩ 

CheckCreate setupone scopeDID-owned forAtlas incident watch with a futureserver-to-server X-Agent-Key header and explicit confirmation. Supply one country or domain watch. This, read-onlya toolpublic neverHTTPS subscribeswebhook, sendsand a caller-generated HMAC key; the key and webhook, orare reportsnever returned. A bare scope without the header remains a read-only setup check. A stored watch asdoes active.not prove alert delivery.

 ⟨6 unchanged words⟩ ":["domain"]}],"properties":{"confirm":{"const":true,"description":"Required true to create a persisted watch","type":"boolean"},"country_code":{"description":"Uppercase ISO 3166-1 alpha-2 ⟨11 unchanged words⟩ ":"DNS domain to assess for watch support","type":"string"},"hmac_key_hex":{"description":"Caller-generated 32-byte webhook verification key; retained by caller","pattern":"^[0-9a-f]{64}$","type":"string"},"min_severity":{"enum":["low","medium","high","critical"],"type":"string"},"webhook_url":{"description":"Public HTTPS webhook destination for signed incident-created events","format":"uri","type":"string"}},"type":"object"}

{"destructiveHint":false,"idempotentHint":true,"openWorldHint":false,"readOnlyHint":truefalse,"title":"WatchCreate SetupAtlas ScopeIncident Watch"}