Country profile · MR
A closer look at internet access.
The score, the signals, the source.
7-day index score
Unavailable
No publishable score in this snapshot. Missing coverage does not establish freedom.
Snapshot fetched
Outside the scored OONI profile
Available source records for Mauritania are shown with their dates and measurement limits. An absent score does not imply unrestricted internet access.
This source branch has no complete scored OONI editorial profile. Current incident and model sources remain independently available.
The original site falls back to the country API and incident records for Mauritania. No unique OONI editorial narrative is available in that branch.
Editorial text is preserved as source context. Product recommendations do not guarantee unrestricted access. The source tier may differ from the measured score because the two have different definitions.
Score window: Unavailable days. Snapshot fetched . Content first seen . Upstream serve-time . These timestamps do not date the all-time OONI observations.
All-time measurements accumulate from an older source file without an observation window or timestamp. The all-time blocked ratio is not the confidence-shrunk score. Zero counts and absent fields are kept distinct.
All-time measurement archive
No historical protocol breakdown in this snapshot.
No collection date or complete-country coverage claim is attached to this archive. Protocol fields can be incomplete and need not sum to the historical total.
Country incident records
ooni · MR-2026-0007
3 source measurements · incident confidence 77.0%
OONI detected 3 days of elevated network interference averaging 100.0% anomaly rate.
ioda · MR-2026-0006
1 source measurements · incident confidence 65.0%
IODA detected significant connectivity drop in MR. 1 critical alerts recorded.
ioda · MR-2026-0005
1 source measurements · incident confidence 65.0%
IODA detected significant connectivity drop in MR. 1 critical alerts recorded.
ioda · MR-2026-0004
1 source measurements · incident confidence 65.0%
IODA detected significant connectivity drop in MR. 1 critical alerts recorded.
ioda · MR-2026-0003
1 source measurements · incident confidence 65.0%
IODA detected significant connectivity drop in MR. 1 critical alerts recorded.
ioda · MR-2026-0002
1 source measurements · incident confidence 65.0%
IODA detected significant connectivity drop in MR. 1 critical alerts recorded.
ioda · MR-2026-0001
1 source measurements · incident confidence 65.0%
IODA detected significant connectivity drop in MR. 1 critical alerts recorded.
Statuses come from each record; “active” is not inferred from severity or an absent end date. Incident confidence is a source field, not independent certainty that censorship occurred.
This is the source mix in the latest returned incident sample, not independent corroboration of the country score. A record can name multiple sources; the counts can overlap.
Dataset version: 2026.11.05. Malformed rows omitted: 0. All readable returned records appear above; each links to its complete evidence and citation page.
Measurement context
This fallback averages up to 50 incident anomaly fields and sums their source measurement counts. It can combine IODA and other signals; it is not an OONI sample block rate or an independently comparable country score.
Source generated: . This is a source timestamp, not proof that every underlying observation is current.
{
"@context": "https://schema.org",
"@type": "Country",
"name": "Mauritania",
"identifier": "MR",
"url": "https://voidly.ai/censorship-index/mr",
"additionalProperty": [
{
"@type": "PropertyValue",
"name": "Censorship Score",
"value": 43,
"description": "0-100 scale where 100 = total internet censorship"
},
{
"@type": "PropertyValue",
"name": "Anomaly Rate",
"value": 0.42747666846504057
},
{
"@type": "PropertyValue",
"name": "Measurement Count",
"value": 9
}
],
"data": {
"name": "Mauritania",
"country": "MR",
"ooni": {
"anomalyRate": 0.42747666846504057,
"measurementCount": 9
},
"incidents": [
{
"affectedAsns": [],
"affectedDomains": [],
"affectedServices": [
"WhatsApp"
],
"anomalyRate": 1,
"confidence": 0.77,
"country": "MR",
"countryName": "Mauritania",
"createdAt": "2026-10-30T10:08:59.312004",
"description": "OONI detected 3 days of elevated network interference averaging 100.0% anomaly rate.",
"duration": 10440,
"durationHours": 174,
"endTime": "2026-10-29T18:00:00Z",
"flag": "🇲🇷",
"hashId": "34d569140ba9",
"id": "MR-2026-0007",
"incidentType": "censorship",
"measurementCount": 3,
"mechanism": null,
"readableId": "MR-2026-0007",
"reportUrl": "https://api.voidly.ai/data/incidents/34d569140ba9/report?format=markdown",
"revision": 1,
"severity": "critical",
"severityGrade": "B",
"sources": [
"ooni"
],
"startTime": "2026-10-22T12:00:00Z",
"status": "confirmed",
"title": "Sustained censorship in MR (2026-10)",
"updatedAt": "2026-10-30T10:08:59.312004"
},
{
"affectedAsns": [],
"affectedDomains": [],
"affectedServices": [],
"anomalyRate": 0.640625,
"confidence": 0.65,
"country": "MR",
"countryName": "Mauritania",
"createdAt": "2026-10-18T06:00:50.991601",
"description": "IODA detected significant connectivity drop in MR. 1 critical alerts recorded.",
"duration": null,
"durationHours": 0,
"endTime": null,
"flag": "🇲🇷",
"hashId": "9a63ef0c7b8b",
"id": "MR-2026-0006",
"incidentType": "disruption",
"measurementCount": 1,
"mechanism": "Infrastructure outage",
"readableId": "MR-2026-0006",
"reportUrl": "https://api.voidly.ai/data/incidents/9a63ef0c7b8b/report?format=markdown",
"revision": 1,
"severity": "critical",
"severityGrade": "B",
"sources": [
"ioda"
],
"startTime": "2026-10-18T00:00:00Z",
"status": "active",
"title": "Internet connectivity disruption in MR",
"updatedAt": "2026-10-18T06:00:50.991601"
},
{
"affectedAsns": [],
"affectedDomains": [],
"affectedServices": [],
"anomalyRate": 0.013953488372093023,
"confidence": 0.65,
"country": "MR",
"countryName": "Mauritania",
"createdAt": "2026-09-28T12:00:08.738035",
"description": "IODA detected significant connectivity drop in MR. 1 critical alerts recorded.",
"duration": 25,
"durationHours": 0.4167,
"endTime": "2026-09-28T11:50:00Z",
"flag": "🇲🇷",
"hashId": "4503cb3892cd",
"id": "MR-2026-0005",
"incidentType": "disruption",
"measurementCount": 1,
"mechanism": "Infrastructure outage",
"readableId": "MR-2026-0005",
"reportUrl": "https://api.voidly.ai/data/incidents/4503cb3892cd/report?format=markdown",
"revision": 2,
"severity": "critical",
"severityGrade": "A",
"sources": [
"ioda"
],
"startTime": "2026-09-28T11:25:00Z",
"status": "active",
"title": "Internet connectivity disruption in MR",
"updatedAt": "2026-09-28T18:00:09.330613"
},
{
"affectedAsns": [],
"affectedDomains": [],
"affectedServices": [],
"anomalyRate": 0.018518518518518517,
"confidence": 0.65,
"country": "MR",
"countryName": "Mauritania",
"createdAt": "2026-09-27T06:00:09.488189",
"description": "IODA detected significant connectivity drop in MR. 1 critical alerts recorded.",
"duration": 1050,
"durationHours": 17.5,
"endTime": "2026-09-27T17:40:00Z",
"flag": "🇲🇷",
"hashId": "df556277e4c6",
"id": "MR-2026-0004",
"incidentType": "disruption",
"measurementCount": 1,
"mechanism": "Infrastructure outage",
"readableId": "MR-2026-0004",
"reportUrl": "https://api.voidly.ai/data/incidents/df556277e4c6/report?format=markdown",
"revision": 3,
"severity": "critical",
"severityGrade": "A",
"sources": [
"ioda"
],
"startTime": "2026-09-27T00:10:00Z",
"status": "active",
"title": "Internet connectivity disruption in MR",
"updatedAt": "2026-09-28T00:00:08.861717"
},
{
"affectedAsns": [],
"affectedDomains": [],
"affectedServices": [],
"anomalyRate": 0.018518518518518517,
"confidence": 0.65,
"country": "MR",
"countryName": "Mauritania",
"createdAt": "2026-09-27T00:15:12.337181",
"description": "IODA detected significant connectivity drop in MR. 1 critical alerts recorded.",
"duration": null,
"durationHours": 0,
"endTime": null,
"flag": "🇲🇷",
"hashId": "e484744ac655",
"id": "MR-2026-0003",
"incidentType": "disruption",
"measurementCount": 1,
"mechanism": "Infrastructure outage",
"readableId": "MR-2026-0003",
"reportUrl": "https://api.voidly.ai/data/incidents/e484744ac655/report?format=markdown",
"revision": 1,
"severity": "critical",
"severityGrade": "A",
"sources": [
"ioda"
],
"startTime": "2026-09-26T23:40:00Z",
"status": "active",
"title": "Internet connectivity disruption in MR",
"updatedAt": "2026-09-27T00:15:12.337181"
},
{
"affectedAsns": [],
"affectedDomains": [],
"affectedServices": [],
"anomalyRate": 0.546875,
"confidence": 0.65,
"country": "MR",
"countryName": "Mauritania",
"createdAt": "2026-06-20T18:00:15.148880",
"description": "IODA detected significant connectivity drop in MR. 1 critical alerts recorded.",
"duration": null,
"durationHours": 0,
"endTime": null,
"flag": "🇲🇷",
"hashId": "ae9fdaada26a",
"id": "MR-2026-0002",
"incidentType": "disruption",
"measurementCount": 1,
"mechanism": "Infrastructure outage",
"readableId": "MR-2026-0002",
"reportUrl": "https://api.voidly.ai/data/incidents/ae9fdaada26a/report?format=markdown",
"revision": 1,
"severity": "critical",
"severityGrade": "B",
"sources": [
"ioda"
],
"startTime": "2026-06-20T11:40:00Z",
"status": "active",
"title": "Internet connectivity disruption in MR",
"updatedAt": "2026-06-20T18:00:15.148880"
},
{
"affectedAsns": [],
"affectedDomains": [],
"affectedServices": [],
"anomalyRate": 0.7538461538461538,
"confidence": 0.65,
"country": "MR",
"countryName": "Mauritania",
"createdAt": "2026-05-04T00:00:11.485768",
"description": "IODA detected significant connectivity drop in MR. 1 critical alerts recorded.",
"duration": null,
"durationHours": 0,
"endTime": null,
"flag": "🇲🇷",
"hashId": "5ff4aecf08f4",
"id": "MR-2026-0001",
"incidentType": "disruption",
"measurementCount": 1,
"mechanism": "Infrastructure outage",
"readableId": "MR-2026-0001",
"reportUrl": "https://api.voidly.ai/data/incidents/5ff4aecf08f4/report?format=markdown",
"revision": 1,
"severity": "critical",
"severityGrade": "B",
"sources": [
"ioda"
],
"startTime": "2026-05-03T20:50:00Z",
"status": "active",
"title": "Internet connectivity disruption in MR",
"updatedAt": "2026-05-04T00:00:11.485768"
}
],
"incidentCount": 7,
"sources": [
"ooni",
"ioda"
],
"affectedServices": [],
"status": "medium",
"score": 43
},
"dataSource": "incident_data",
"source": "Voidly Global Censorship Index",
"lastUpdated": "2026-11-05T23:10:44.364Z"
}Quality measures evidence confidence, not censorship probability. A small or absent sample cannot establish country-wide access. Quality generated: .
Quality sourceProbe network statistics are not published while we review what they reveal. No country-origin probe count is shown.
Probe aggregates, per-probe and operator listings are temporarily withheld to protect volunteer operators. Historical country-origin records, if returned, do not establish current or whole-country coverage.
Model outlook
7-day event probability
Unavailable
A censorship or connectivity event within seven days of the model’s as-of date, relative to the country’s baseline. It does not predict an exact shutdown date or measure standing freedom.
Model: v1
Model as-of date: Date unavailable
Source generated: . This is a source timestamp, not proof that every underlying observation is current.
fixed in-sample model confidence, NOT a per-forecast calibrated probability
Use for cross-country RISK RANKING, not precise within-country day-of-shutdown timing.
This forecast scores the ONSET of a new recorded incident, not how censored a country already is. A permanently and stably blocked network generates no new anomaly, so it scores LOW. Measured 2026-07-31: China 0.015 and India 0.023 both rank BELOW Denmark 0.062 — China is probed 30/30 days across 836 domains and has logged nothing for 71 days, while Denmark is probed 2/30 days. A low score here means 'no CHANGE expected', never 'this country is free'. For standing severity use /v1/atlas/score-v2 or the censorship index instead.
Training ROC AUC 0.954 came from a RANDOM stratified split (consecutive country-days in an incident window are near-duplicates, inflating it). No target leakage: features use [date-7,date], label uses [date+1,date+7].
~0.85-0.90 (dual-holdout temporal gate)
ML_LEAKAGE_AUDIT.md (forecast v1/7day row).
{
"aci": {
"active": false,
"reason": "unverified: training holdout and ACI outcomes do not match the served headline or qualified target"
},
"aci_alpha": null,
"confidence": 0.85,
"country": "MR",
"country_name": "MR",
"covered": true,
"forecast": [
{
"date": "2026-11-05",
"day": 0,
"drivers": [],
"risk": 0.02
},
{
"date": "2026-11-06",
"day": 1,
"drivers": [],
"risk": 0.01
},
{
"date": "2026-11-07",
"day": 2,
"drivers": [],
"risk": 0.022
},
{
"date": "2026-11-08",
"day": 3,
"drivers": [],
"risk": 0.032
},
{
"date": "2026-11-09",
"day": 4,
"drivers": [],
"risk": 0.01
},
{
"date": "2026-11-10",
"day": 5,
"drivers": [],
"risk": 0.019
},
{
"date": "2026-11-11",
"day": 6,
"drivers": [],
"risk": 0.049
},
{
"date": "2026-11-12",
"day": 7,
"drivers": [],
"risk": 0.043
}
],
"generated_at": "2026-11-05T23:10:44.148244",
"honest_caveat": {
"confidence_is": "fixed in-sample model confidence, NOT a per-forecast calibrated probability",
"correct_use": "Use for cross-country RISK RANKING, not precise within-country day-of-shutdown timing.",
"detects_change_not_state": "This forecast scores the ONSET of a new recorded incident, not how censored a country already is. A permanently and stably blocked network generates no new anomaly, so it scores LOW. Measured 2026-07-31: China 0.015 and India 0.023 both rank BELOW Denmark 0.062 — China is probed 30/30 days across 836 domains and has logged nothing for 71 days, while Denmark is probed 2/30 days. A low score here means 'no CHANGE expected', never 'this country is free'. For standing severity use /v1/atlas/score-v2 or the censorship index instead.",
"headline_auc_note": "Training ROC AUC 0.954 came from a RANDOM stratified split (consecutive country-days in an incident window are near-duplicates, inflating it). No target leakage: features use [date-7,date], label uses [date+1,date+7].",
"honest_temporal_holdout_auc": "~0.85-0.90 (dual-holdout temporal gate)",
"live_forward_accountability": [
"GET /v1/sentinel/accuracy (status unvalidated; empirical precision/recall withheld)",
"GET /v1/forecast/onset-skill (honest onset / forward skill)"
],
"source": "ML_LEAKAGE_AUDIT.md (forecast v1/7day row)."
},
"interval_90": null,
"model_version": "v1",
"summary": {
"avg_risk": 0.026,
"key_drivers": [],
"max_risk": 0.049,
"max_risk_day": 6
},
"top_features": [
{
"contribution": -0.0805,
"direction": "down",
"name": "week_of_year",
"source": "model"
},
{
"contribution": -0.0478,
"direction": "down",
"name": "gdelt_unrest_30d",
"source": "model"
},
{
"contribution": -0.015,
"direction": "down",
"name": "block_rate_roll30_mean",
"source": "model"
}
],
"top_features_per_day": [
{
"base_prob": 0,
"date": "2026-11-05",
"day": 0,
"risk": 0.02,
"top_features": [
{
"contribution": -0.0805,
"direction": "down",
"name": "week_of_year",
"source": "model"
},
{
"contribution": -0.0478,
"direction": "down",
"name": "gdelt_unrest_30d",
"source": "model"
},
{
"contribution": -0.015,
"direction": "down",
"name": "block_rate_roll30_mean",
"source": "model"
}
]
},
{
"base_prob": 0,
"date": "2026-11-06",
"day": 1,
"risk": 0.01,
"top_features": [
{
"contribution": -0.0805,
"direction": "down",
"name": "week_of_year",
"source": "model"
},
{
"contribution": -0.0478,
"direction": "down",
"name": "gdelt_unrest_30d",
"source": "model"
},
{
"contribution": -0.015,
"direction": "down",
"name": "block_rate_roll30_mean",
"source": "model"
}
]
},
{
"base_prob": 0,
"date": "2026-11-07",
"day": 2,
"risk": 0.022,
"top_features": [
{
"contribution": -0.0805,
"direction": "down",
"name": "week_of_year",
"source": "model"
},
{
"contribution": -0.0478,
"direction": "down",
"name": "gdelt_unrest_30d",
"source": "model"
},
{
"contribution": -0.015,
"direction": "down",
"name": "block_rate_roll30_mean",
"source": "model"
}
]
},
{
"base_prob": 0,
"date": "2026-11-08",
"day": 3,
"risk": 0.032,
"top_features": [
{
"contribution": -0.0805,
"direction": "down",
"name": "week_of_year",
"source": "model"
},
{
"contribution": -0.0478,
"direction": "down",
"name": "gdelt_unrest_30d",
"source": "model"
},
{
"contribution": -0.015,
"direction": "down",
"name": "block_rate_roll30_mean",
"source": "model"
}
]
},
{
"base_prob": 0,
"date": "2026-11-09",
"day": 4,
"risk": 0.01,
"top_features": [
{
"contribution": -0.0805,
"direction": "down",
"name": "week_of_year",
"source": "model"
},
{
"contribution": -0.0478,
"direction": "down",
"name": "gdelt_unrest_30d",
"source": "model"
},
{
"contribution": 0.02,
"direction": "up",
"name": "day_decay_t+4",
"source": "overlay"
}
]
},
{
"base_prob": 0,
"date": "2026-11-10",
"day": 5,
"risk": 0.019,
"top_features": [
{
"contribution": -0.0805,
"direction": "down",
"name": "week_of_year",
"source": "model"
},
{
"contribution": -0.0478,
"direction": "down",
"name": "gdelt_unrest_30d",
"source": "model"
},
{
"contribution": 0.025,
"direction": "up",
"name": "day_decay_t+5",
"source": "overlay"
}
]
},
{
"base_prob": 0,
"date": "2026-11-11",
"day": 6,
"risk": 0.049,
"top_features": [
{
"contribution": -0.0805,
"direction": "down",
"name": "week_of_year",
"source": "model"
},
{
"contribution": -0.0478,
"direction": "down",
"name": "gdelt_unrest_30d",
"source": "model"
},
{
"contribution": 0.03,
"direction": "up",
"name": "day_decay_t+6",
"source": "overlay"
}
]
},
{
"base_prob": 0,
"date": "2026-11-12",
"day": 7,
"risk": 0.043,
"top_features": [
{
"contribution": -0.0805,
"direction": "down",
"name": "week_of_year",
"source": "model"
},
{
"contribution": -0.0478,
"direction": "down",
"name": "gdelt_unrest_30d",
"source": "model"
},
{
"contribution": 0.035,
"direction": "up",
"name": "day_decay_t+7",
"source": "overlay"
}
]
}
]
}The returned daily shape belongs to legacy v1. It is not independently validated and does not describe the honest 7-day headline model above. Its peak day, fixed confidence and interval must not be attached to that headline.
| Date | Legacy value | Source drivers |
|---|---|---|
| 2026-11-05 | 2.0% | |
| 2026-11-06 | 1.0% | |
| 2026-11-07 | 2.2% | |
| 2026-11-08 | 3.2% | |
| 2026-11-09 | 1.0% | |
| 2026-11-10 | 1.9% | |
| 2026-11-11 | 4.9% | |
| 2026-11-12 | 4.3% |
Original display confidence field: 85.0%. The source describes this as fixed in-sample model confidence, not calibrated per-forecast certainty.
Separate multi-horizon model. Each mark is a model score. The former residual bands lack validated 90% coverage and are withheld.
22.1% · interval unvalidated
58.8% · interval unvalidated
70.0% · interval unvalidated
Model: multi-horizon-h-v1. Interval coverage is unvalidated; these estimates do not support precise within-country timing.
Source consistency: within the model’s monotonicity tolerance.
Per-horizon breakdownSource generated: . This is a source timestamp, not proof that every underlying observation is current.
AUC (e.g. 1d ~0.91) OVERSTATES usefulness — per-horizon F1 is much lower (~0.45 at 1d) because shutdown-positive days are rare and the 7/30-day target is a sliding window that is highly autocorrelated day-to-day (a leakage-prone label).
LOCO binary-outcome residual widths are not validated 90% probability intervals; bands are withheld at every horizon.
Use for cross-country RISK RANKING within the watched panel, NOT precise within-country day-of-shutdown timing. Data-backed for covered countries only (see the 'covered' flag).
This forecast scores the ONSET of a new recorded incident, not how censored a country already is. A permanently and stably blocked network generates no new anomaly, so it scores LOW. Measured 2026-07-31: China 0.015 and India 0.023 both rank BELOW Denmark 0.062 — China is probed 30/30 days across 836 domains and has logged nothing for 71 days, while Denmark is probed 2/30 days. A low score here means 'no CHANGE expected', never 'this country is free'. For standing severity use /v1/atlas/score-v2 or the censorship index instead.
Per-horizon AUC is leave-one-country-out on the SPOTLIGHT (watched) panel, NOT all 130 countries; see /v1/forecast/multi-horizon/info for the per-horizon F1.
{
"consistency": {
"monotonic": true,
"violations": []
},
"country": "MR",
"country_name": "MR",
"covered": true,
"generated_at": "2026-11-05T23:10:43.838638",
"honest_caveat": {
"auc_caveat": "AUC (e.g. 1d ~0.91) OVERSTATES usefulness — per-horizon F1 is much lower (~0.45 at 1d) because shutdown-positive days are rare and the 7/30-day target is a sliding window that is highly autocorrelated day-to-day (a leakage-prone label).",
"conformal_caveat": "LOCO binary-outcome residual widths are not validated 90% probability intervals; bands are withheld at every horizon.",
"correct_use": "Use for cross-country RISK RANKING within the watched panel, NOT precise within-country day-of-shutdown timing. Data-backed for covered countries only (see the 'covered' flag).",
"detects_change_not_state": "This forecast scores the ONSET of a new recorded incident, not how censored a country already is. A permanently and stably blocked network generates no new anomaly, so it scores LOW. Measured 2026-07-31: China 0.015 and India 0.023 both rank BELOW Denmark 0.062 — China is probed 30/30 days across 836 domains and has logged nothing for 71 days, while Denmark is probed 2/30 days. A low score here means 'no CHANGE expected', never 'this country is free'. For standing severity use /v1/atlas/score-v2 or the censorship index instead.",
"metrics_are": "Per-horizon AUC is leave-one-country-out on the SPOTLIGHT (watched) panel, NOT all 130 countries; see /v1/forecast/multi-horizon/info for the per-horizon F1."
},
"honest_horizons": [
"1d",
"30d",
"7d"
],
"horizons": {
"1d": {
"conformal_90": null,
"horizon": "1d",
"horizon_days": 1,
"interval_status": "unvalidated",
"probability": 0.221,
"top_features": [
{
"feature": "week_of_year",
"shap_value": -0.9548981189727783
},
{
"feature": "block_rate_roll7_std",
"shap_value": -0.12384400516748428
},
{
"feature": "block_rate_roll7_mean",
"shap_value": -0.09496477246284485
},
{
"feature": "block_rate_roll30_mean",
"shap_value": -0.04210022836923599
},
{
"feature": "gdelt_unrest_30d",
"shap_value": -0.03528587147593498
}
]
},
"30d": {
"conformal_90": null,
"horizon": "30d",
"horizon_days": 30,
"interval_status": "unvalidated",
"probability": 0.6999,
"top_features": [
{
"feature": "week_of_year",
"shap_value": -0.7989568710327148
},
{
"feature": "block_rate_roll7_std",
"shap_value": -0.1609630286693573
},
{
"feature": "risk_tier",
"shap_value": -0.13833554089069366
},
{
"feature": "recent_shutdown",
"shap_value": -0.10059788823127747
},
{
"feature": "block_rate_lag1",
"shap_value": -0.0639578253030777
}
]
},
"7d": {
"conformal_90": null,
"horizon": "7d",
"horizon_days": 7,
"interval_status": "unvalidated",
"probability": 0.5884,
"top_features": [
{
"feature": "week_of_year",
"shap_value": -0.7154033184051514
},
{
"feature": "block_rate_roll7_std",
"shap_value": -0.2002677321434021
},
{
"feature": "recent_shutdown",
"shap_value": 0.13047538697719574
},
{
"feature": "month",
"shap_value": -0.07534761726856232
},
{
"feature": "gdelt_unrest_30d",
"shap_value": -0.0722813680768013
}
]
}
},
"model_version": "multi-horizon-h-v1"
}Model evaluation and labels differ. Honest 7-day model · Multi-horizon evaluation. An API response date is not the date of the underlying training data.
Additional signals
Each source answers a different question. Missing signals stay visible.
Unavailable
Separate from the index score. A weighted signal composite; sparse evidence can pull it toward zero.
Source freshness is unknown.
Date unavailable
Score methodSource generated: Date unavailable. This is a source timestamp, not proof that every underlying observation is current.
202 median days
Scenario: a critical censorship incident in Mauritania. This is not a timer for an existing incident.
Source interquartile range: 5–401 days.
Exploratory · source model is below its promotion threshold.
Model trained: . Censoring in training data: 78.4%.
Source generated: Date unavailable. This is a source timestamp, not proof that every underlying observation is current.
This read-only inference endpoint requires a POST request:
POST https://api.voidly.ai/v1/forecast/duration
Content-Type: application/json
{
"country": "MR",
"severity": "critical",
"incident_type": "censorship"
}{
"country": "MR",
"features_used": {
"continent_Africa": 0,
"continent_Americas": 0,
"continent_Asia": 0,
"continent_Europe": 0,
"continent_Oceania": 0,
"country_risk_tier": 3,
"first_seen_month": 11,
"first_seen_year": 2026,
"prior_shutdowns_count_24mo": 1,
"severity_critical": 1,
"severity_warning": 0,
"type_censorship": 1,
"type_mixed": 0
},
"honest_caveats": [
"country not found in country_geography — using 'no continent' encoding",
"model below promote floor (c=0.55, n=343) — treat as exploratory",
"censoring rate in training set: 78%"
],
"incident_type": "censorship",
"iqr_days": 396,
"median_days": 202,
"median_hours": 4848,
"model": {
"censoring_rate": 0.7842565597667639,
"class": "RandomSurvivalForest",
"feature_names": [
"country_risk_tier",
"first_seen_year",
"first_seen_month",
"prior_shutdowns_count_24mo",
"continent_Africa",
"continent_Americas",
"continent_Asia",
"continent_Europe",
"continent_Oceania",
"severity_critical",
"severity_warning",
"type_censorship",
"type_mixed"
],
"name": "shutdown_duration_rsf_v1",
"trained_at": "2026-05-21T12:01:30.451711Z"
},
"model_c_index": 0.5455,
"n_training_samples": 343,
"p25_hours": 120,
"p75_hours": 9624,
"promoted": false,
"severity": "critical",
"survival_curve_summary": [
{
"hours": 120,
"survival_prob": 0.9
},
{
"hours": 120,
"survival_prob": 0.75
},
{
"hours": 4848,
"survival_prob": 0.5
},
{
"hours": 9624,
"survival_prob": 0.25
},
{
"hours": 17880,
"survival_prob": 0.1
}
]
}Follower estimate unavailable.
Trigger attribution is probabilistic, not causal. Noisy-OR aggregation can overstate risk when triggers share a cause.
Source generated: Date unavailable. This is a source timestamp, not proof that every underlying observation is current.
Similarity of measured blocking patterns, not governments or intent. Sparse feature coverage lowers confidence.
Source generated: . This is a source timestamp, not proof that every underlying observation is current.
Similarity here is BEHAVIORAL, not political. Two countries are near each other because they BLOCK in similar ways — comparable method mix, target categories and temporal shape — which does NOT imply comparable governments, laws or intent. A stable democracy with seasonal sports-piracy blocks can sit near an authoritarian state if their measured blocking footprints rhyme. Feature vectors are SPARSE for low-coverage countries (flagged sparse=true, <3 of 5 behavioral feature families populated) — their neighbors are dominated by regime/continent/cohort one-hots and should be read with low confidence. The 2D graph projection (UMAP or PCA) compresses a 32-dimensional space to 2 and loses information — use the cosine `neighbors` list, not pixel distance in the projection, for any quantitative claim.
{
"country": "MR",
"generated_at": "2026-11-02T05:10:36.146797+00:00",
"honest_caveat": "Similarity here is BEHAVIORAL, not political. Two countries are near each other because they BLOCK in similar ways — comparable method mix, target categories and temporal shape — which does NOT imply comparable governments, laws or intent. A stable democracy with seasonal sports-piracy blocks can sit near an authoritarian state if their measured blocking footprints rhyme. Feature vectors are SPARSE for low-coverage countries (flagged sparse=true, <3 of 5 behavioral feature families populated) — their neighbors are dominated by regime/continent/cohort one-hots and should be read with low confidence. The 2D graph projection (UMAP or PCA) compresses a 32-dimensional space to 2 and loses information — use the cosine `neighbors` list, not pixel distance in the projection, for any quantitative claim.",
"method": "standardized-cosine + pca",
"nearest": [
{
"continent": null,
"country": "LR",
"risk_tier": null,
"similarity": 0.9298,
"sparse": false
},
{
"continent": null,
"country": "SZ",
"risk_tier": null,
"similarity": 0.9152,
"sparse": false
},
{
"continent": null,
"country": "TL",
"risk_tier": null,
"similarity": 0.9117,
"sparse": false
},
{
"continent": null,
"country": "MU",
"risk_tier": null,
"similarity": 0.8952,
"sparse": false
},
{
"continent": null,
"country": "BJ",
"risk_tier": null,
"similarity": 0.8914,
"sparse": false
},
{
"continent": null,
"country": "BZ",
"risk_tier": null,
"similarity": 0.848,
"sparse": false
},
{
"continent": null,
"country": "SL",
"risk_tier": null,
"similarity": 0.8323,
"sparse": false
},
{
"continent": null,
"country": "NA",
"risk_tier": null,
"similarity": 0.8221,
"sparse": false
},
{
"continent": null,
"country": "LS",
"risk_tier": null,
"similarity": 0.809,
"sparse": false
},
{
"continent": null,
"country": "GY",
"risk_tier": null,
"similarity": 0.805,
"sparse": false
}
],
"query_meta": {
"continent": null,
"country": "MR",
"dtw_cohort": null,
"evidence_count": 27,
"feature_families_present": 3,
"feature_families_total": 5,
"risk_tier": null,
"sparse": false
},
"version": "v1"
}Grades are source rubric categories. They do not independently establish deliberate blocking.
Grading rubricSource generated: Date unavailable. This is a source timestamp, not proof that every underlying observation is current.
{
"distribution": {
"A": 3,
"B": 4,
"C": 0,
"F": 0
},
"filters": {
"country": "MR",
"incident_type": null
},
"rules": {
"A": "else",
"B": "n_sources >= 1 AND avg_confidence >= 0.7",
"C": "n_sources >= 2 AND n_domains >= 3",
"F": "duration_hours >= 24 AND n_sources >= 2"
},
"total": 7,
"ungraded": 0
}No sufficiently covered headline recommendation is available.
Source generated: . This is a source timestamp, not proof that every underlying observation is current.
{
"country_code": "MR",
"do_not_share_card_without_caveats": true,
"endpoints": {
"info": "/v1/atlas/circumvention/info",
"upstream_evasion_table": "/v1/atlas/evasion/MR"
},
"generated_at": "2026-11-05T06:00:11.178847+00:00",
"honest_caveats": [
"PROBE SUCCESS != REAL-USER SUCCESS. DPI vendors increasingly rate-limit real client traffic while letting OONI / Voidly synthetic measurement probes through. Success rates here are an UPPER BOUND on what a real user will see.",
"Recommendations can become stale within hours. Censorship adapts. If a try_first tool stops working, the next probe rotation (cron daily 06:00 UTC) will catch it but you will see it before we do.",
"'Try first' is NOT 'safe'. It means 'highest measured success rate, highest probe confidence.' Operational security (unique device, throwaway account, no SIM, no DNS leak, etc.) is the USER's responsibility — none of that is captured here.",
"Domain-fronted tools (Snowflake, Lantern, Psiphon) are SYSTEMATICALLY UNDER-ESTIMATED. Bootstrap-domain probes do not measure the rotating tunnel endpoints those tools actually use.",
"Self-hosted WireGuard / OpenVPN on a private IP is INVISIBLE to OONI probes and is not represented here. If you have one, trust it over any commercial recommendation.",
"Tor rows roll up vanilla / obfs4 / Snowflake / Meek transports. Per-transport granularity needs raw OONI nettests, out of scope for v1.",
"DO NOT share this recommendation as a forum-shareable card without the caveat block. The do_not_share_card_without_caveats flag in the response is on by design."
],
"informational_disclaimer": "This endpoint is informational, not a safety guarantee. Do not share as a forum-shareable card without the caveat block.",
"insufficient_coverage": true,
"lookback_days": 30,
"n_tools_measured": 0,
"n_total_probes": 0,
"note": "No circumvention-tool probe measurements found for MR in the last 30 days. This usually means OONI probe coverage is absent in MR, NOT that every tool works. We cannot recommend anything for a country we cannot measure.",
"schema": "voidly-circumvention-recommendation/v1",
"tiers": {
"avoid": [],
"fallback": [],
"try_first": []
},
"tools_ranked": []
}Editorial service list
Undated editorial context. Check current evidence before treating a service as blocked.
No editorial blocked-service list is available in this source branch.
Domain-check status and limitsCountry guidance
The original country recommendations are preserved below. They are undated guidance, not a guarantee of access or safety.
Questions, answered
This dated index does not have a publishable score for this country. Missing measurement coverage is not evidence of unrestricted access.
Use the recent records and their original sources to investigate a particular restriction. Disruption alone does not establish intentional censorship.
There is no scored OONI editorial profile in this source branch. Use the dated incident evidence and domain checks for specific services.
Results can differ between networks, measurement times and targets. Country incident records and the country API preserve source details.
The index score snapshot was fetched . Its source serve-time can advance without changes in the underlying scores. The accumulating historical OONI file has no dated observation window.
Current country data, incidents, network state and models have independent source dates. The page read those responses . Public GET reads request five-minute revalidation; response read time is not observation time. Model POST requests remain uncached. This page does not continuously refresh. Missing, stale and uncovered responses remain labeled.
Subscribe to source feedsDirect domain-check output is temporarily withheld while source-safety review is in progress. The web checker, API, MCP and CLI do not currently provide a fresh check here.
Use dated country incident evidence to inspect reported domains, observation times and methods. Historical evidence is not a current availability result or a guarantee for every user.
Use the canonical country URL, https://voidly.ai/mr, with your access date and the date of any source snapshot you quote. The citation below supports plain text and BibTeX copy/download. A model generation or page access date is not a publication date.
Voidly-original analysis is CC BY 4.0. Upstream records keep their own attribution and licenses.
Open evidence
Voidly Research. (n.d.). Mauritania Internet Censorship Profile. https://voidly.ai/mr (accessed 2026-11-05).
Citations use the public /mr URL. Access date is not publication date. Cite the individual source, observation time and model version when using a number.
Voidly-original analysis and annotations are CC BY 4.0. Upstream OONI, IODA and Censored Planet records retain their own licenses. Read and attribute the underlying evidence.
Agent summary
Facts: Mauritania: dated index score unavailable; unavailable measurements in that score window. The sample does not establish a domain or service block verdict.
As of score observation: Unavailable. Score snapshot fetched: 2026-11-04T15:49:45.946Z. Fetch time is not the underlying measurement time. OONI source last fetched: unavailable; this is a separate source fetch, not an observation time.
Sources: 2 distinct names (ooni, ioda) in 7 returned incident records, up to 10. Missing source lists: 0. This is not the score source count or independent corroboration.
Cite: Voidly Research. (n.d.). Mauritania Internet Censorship Profile. https://voidly.ai/mr (accessed 2026-11-05).
Use this in your agent
Connect the read-only Atlas MCP service, then select and copy this request into your agent.
Atlas MCP setup · Agent summary JSON · Endpoint: https://atlas-mcp.voidly.ai/mcp