# AI # To Markdown ## Convert Files into Markdown **post** `/accounts/{account_id}/ai/tomarkdown` Converts uploaded files into Markdown format using Workers AI. ### Path Parameters - `account_id: string` ### Returns - `result: array of object { data, format, mimeType, 2 more }` - `data: string` - `format: string` - `mimeType: string` - `name: string` - `tokens: string` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/ai/tomarkdown \ -H 'Content-Type: multipart/form-data' \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \ -F files='["Example data"]' ``` #### Response ```json { "result": [ { "data": "data", "format": "format", "mimeType": "mimeType", "name": "name", "tokens": "tokens" } ], "success": true } ``` ## Domain Types ### To Markdown Create Response - `ToMarkdownCreateResponse object { data, format, mimeType, 2 more }` - `data: string` - `format: string` - `mimeType: string` - `name: string` - `tokens: string` # Inference ## Get Workers AI inference distribution by dimension **get** `/radar/ai/inference/summary/{dimension}` Retrieves an aggregated summary of the number of inferences run on Workers AI, grouped by the specified dimension. ### Path Parameters - `dimension: "MODEL" or "TASK"` Specifies the attribute by which to group the results. - `"MODEL"` - `"TASK"` ### Query Parameters - `asn: optional array of string` Filters results by Autonomous System. Specify one or more Autonomous System Numbers (ASNs) as a comma-separated list. Prefix with `-` to exclude ASNs from results. For example, `-174, 3356` excludes results from AS174, but includes results from AS3356. - `continent: optional array of string` Filters results by continent. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude continents from results. For example, `-EU,NA` excludes results from EU, but includes results from NA. - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `location: optional array of string` Filters results by location. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude locations from results. For example, `-US,PT` excludes results from the US, but includes results from PT. - `name: optional array of string` Array of names used to label the series in the response. ### Returns - `result: object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/inference/summary/$DIMENSION \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "summary_0": { "@cf/meta/llama-3-8b-instruct": "8.381743", "@cf/meta/m2m100-1.2b": "22.904", "@cf/stabilityai/stable-diffusion-xl-base-1.0": "10.274394" } }, "success": true } ``` ## Get time series distribution of Workers AI inference by dimension. **get** `/radar/ai/inference/timeseries_groups/{dimension}` Retrieves the distribution of the number of inferences run on Workers AI, grouped by the specified dimension over time. ### Path Parameters - `dimension: "MODEL" or "TASK"` Specifies the attribute by which to group the results. - `"MODEL"` - `"TASK"` ### Query Parameters - `aggInterval: optional "15m" or "1h" or "1d" or "1w"` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). When omitted, the interval is auto-selected from the requested date range; finer intervals are only available for shorter ranges. If the requested interval is too granular for the date range, the request is rejected. - `"15m"` - `"1h"` - `"1d"` - `"1w"` - `asn: optional array of string` Filters results by Autonomous System. Specify one or more Autonomous System Numbers (ASNs) as a comma-separated list. Prefix with `-` to exclude ASNs from results. For example, `-174, 3356` excludes results from AS174, but includes results from AS3356. - `continent: optional array of string` Filters results by continent. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude continents from results. For example, `-EU,NA` excludes results from EU, but includes results from NA. - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `location: optional array of string` Filters results by location. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude locations from results. For example, `-US,PT` excludes results from the US, but includes results from PT. - `name: optional array of string` Array of names used to label the series in the response. - `normalization: optional "PERCENTAGE" or "MIN0_MAX"` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` ### Returns - `result: object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/inference/timeseries_groups/$DIMENSION \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "aggInterval": "FIFTEEN_MINUTES", "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "serie_0": { "timestamps": [ "2023-08-08T10:15:00Z" ] } }, "success": true } ``` ## Domain Types ### Inference Summary V2 Response - `InferenceSummaryV2Response object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` ### Inference Timeseries Groups V2 Response - `InferenceTimeseriesGroupsV2Response object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` # Summary ## Get Workers AI models summary **get** `/radar/ai/inference/summary/model` Retrieves the distribution of the number of inferences by model. ### Query Parameters - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `name: optional array of string` Array of names used to label the series in the response. ### Returns - `result: object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/inference/summary/model \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "summary_0": { "@cf/meta/llama-3-8b-instruct": "8.381743", "@cf/meta/m2m100-1.2b": "22.904", "@cf/stabilityai/stable-diffusion-xl-base-1.0": "10.274394" } }, "success": true } ``` ## Get Workers AI tasks summary **get** `/radar/ai/inference/summary/task` Retrieves the distribution of the number of inferences by task. ### Query Parameters - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `name: optional array of string` Array of names used to label the series in the response. ### Returns - `result: object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/inference/summary/task \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "summary_0": { "text generation": "10.274394", "text-to-image": "22.904" } }, "success": true } ``` ## Domain Types ### Summary Model Response - `SummaryModelResponse object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` ### Summary Task Response - `SummaryTaskResponse object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` # Timeseries Groups # Summary ## Get Workers AI models time series **get** `/radar/ai/inference/timeseries_groups/model` Retrieves the distribution of the number of inferences by model over time. ### Query Parameters - `aggInterval: optional "15m" or "1h" or "1d" or "1w"` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). When omitted, the interval is auto-selected from the requested date range; finer intervals are only available for shorter ranges. If the requested interval is too granular for the date range, the request is rejected. - `"15m"` - `"1h"` - `"1d"` - `"1w"` - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `name: optional array of string` Array of names used to label the series in the response. ### Returns - `result: object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/inference/timeseries_groups/model \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "aggInterval": "FIFTEEN_MINUTES", "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "serie_0": { "timestamps": [ "2023-08-08T10:15:00Z" ] } }, "success": true } ``` ## Get Workers AI tasks time series **get** `/radar/ai/inference/timeseries_groups/task` Retrieves the distribution of the number of inferences by task over time. ### Query Parameters - `aggInterval: optional "15m" or "1h" or "1d" or "1w"` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). When omitted, the interval is auto-selected from the requested date range; finer intervals are only available for shorter ranges. If the requested interval is too granular for the date range, the request is rejected. - `"15m"` - `"1h"` - `"1d"` - `"1w"` - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `name: optional array of string` Array of names used to label the series in the response. ### Returns - `result: object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/inference/timeseries_groups/task \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "aggInterval": "FIFTEEN_MINUTES", "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "serie_0": { "timestamps": [ "2023-08-08T10:15:00Z" ] } }, "success": true } ``` ## Domain Types ### Summary Model Response - `SummaryModelResponse object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` ### Summary Task Response - `SummaryTaskResponse object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` # Bots ## Get AI bots HTTP requests distribution by dimension **get** `/radar/ai/bots/summary/{dimension}` Retrieves an aggregated summary of AI bots HTTP requests grouped by the specified dimension. ### Path Parameters - `dimension: "USER_AGENT" or "CRAWL_PURPOSE" or "INDUSTRY" or 4 more` Specifies the attribute by which to group the results. - `"USER_AGENT"` - `"CRAWL_PURPOSE"` - `"INDUSTRY"` - `"VERTICAL"` - `"CONTENT_TYPE"` - `"RESPONSE_STATUS"` - `"RESPONSE_STATUS_CATEGORY"` ### Query Parameters - `asn: optional array of string` Filters results by Autonomous System. Specify one or more Autonomous System Numbers (ASNs) as a comma-separated list. Prefix with `-` to exclude ASNs from results. For example, `-174, 3356` excludes results from AS174, but includes results from AS3356. - `contentType: optional array of "HTML" or "IMAGES" or "JSON" or 13 more` Filters results by content type category. When set, results can only be further filtered by location, continent, or Autonomous System. - `"HTML"` - `"IMAGES"` - `"JSON"` - `"JAVASCRIPT"` - `"CSS"` - `"PLAIN_TEXT"` - `"FONTS"` - `"XML"` - `"YAML"` - `"VIDEO"` - `"AUDIO"` - `"MARKDOWN"` - `"DOCUMENTS"` - `"BINARY"` - `"SERIALIZATION"` - `"OTHER"` - `continent: optional array of string` Filters results by continent. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude continents from results. For example, `-EU,NA` excludes results from EU, but includes results from NA. - `crawlPurpose: optional array of string` Filters results by bot crawl purpose. - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `industry: optional array of string` Filters results by industry. - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `location: optional array of string` Filters results by location. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude locations from results. For example, `-US,PT` excludes results from the US, but includes results from PT. - `name: optional array of string` Array of names used to label the series in the response. - `responseStatus: optional array of string` Filters results by HTTP response status code (e.g. 200, 403, 404). Only [IANA-registered codes](https://www.iana.org/assignments/http-status-codes/http-status-codes.xhtml) are accepted. - `responseStatusCategory: optional array of "INFORMATIONAL" or "SUCCESS" or "REDIRECTION" or 2 more` Filters results by HTTP response status code category. - `"INFORMATIONAL"` - `"SUCCESS"` - `"REDIRECTION"` - `"CLIENT_ERROR"` - `"SERVER_ERROR"` - `userAgent: optional array of string` Filters results by user agent. - `vertical: optional array of string` Filters results by vertical. ### Returns - `result: object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/bots/summary/$DIMENSION \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "summary_0": { "Amazonbot": "10.274394", "Bytespider": "8.381743", "facebookexternalhit": "63.40249" } }, "success": true } ``` ## Get AI bots HTTP requests time series **get** `/radar/ai/bots/timeseries` Retrieves AI bots HTTP request volume over time. ### Query Parameters - `aggInterval: optional "15m" or "1h" or "1d" or "1w"` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). When omitted, the interval is auto-selected from the requested date range; finer intervals are only available for shorter ranges. If the requested interval is too granular for the date range, the request is rejected. - `"15m"` - `"1h"` - `"1d"` - `"1w"` - `asn: optional array of string` Filters results by Autonomous System. Specify one or more Autonomous System Numbers (ASNs) as a comma-separated list. Prefix with `-` to exclude ASNs from results. For example, `-174, 3356` excludes results from AS174, but includes results from AS3356. - `contentType: optional array of "HTML" or "IMAGES" or "JSON" or 13 more` Filters results by content type category. When set, results can only be further filtered by location, continent, or Autonomous System. - `"HTML"` - `"IMAGES"` - `"JSON"` - `"JAVASCRIPT"` - `"CSS"` - `"PLAIN_TEXT"` - `"FONTS"` - `"XML"` - `"YAML"` - `"VIDEO"` - `"AUDIO"` - `"MARKDOWN"` - `"DOCUMENTS"` - `"BINARY"` - `"SERIALIZATION"` - `"OTHER"` - `continent: optional array of string` Filters results by continent. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude continents from results. For example, `-EU,NA` excludes results from EU, but includes results from NA. - `crawlPurpose: optional array of string` Filters results by bot crawl purpose. - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `industry: optional array of string` Filters results by industry. - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `location: optional array of string` Filters results by location. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude locations from results. For example, `-US,PT` excludes results from the US, but includes results from PT. - `name: optional array of string` Array of names used to label the series in the response. - `responseStatus: optional array of string` Filters results by HTTP response status code (e.g. 200, 403, 404). Only [IANA-registered codes](https://www.iana.org/assignments/http-status-codes/http-status-codes.xhtml) are accepted. - `responseStatusCategory: optional array of "INFORMATIONAL" or "SUCCESS" or "REDIRECTION" or 2 more` Filters results by HTTP response status code category. - `"INFORMATIONAL"` - `"SUCCESS"` - `"REDIRECTION"` - `"CLIENT_ERROR"` - `"SERVER_ERROR"` - `userAgent: optional array of string` Filters results by user agent. - `vertical: optional array of string` Filters results by vertical. ### Returns - `result: object { meta }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/bots/timeseries \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "aggInterval": "FIFTEEN_MINUTES", "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] } }, "success": true } ``` ## Get time series distribution of AI bots HTTP requests by dimension. **get** `/radar/ai/bots/timeseries_groups/{dimension}` Retrieves the distribution of HTTP requests from AI bots, grouped by the specified dimension over time. ### Path Parameters - `dimension: "USER_AGENT" or "CRAWL_PURPOSE" or "INDUSTRY" or 4 more` Specifies the attribute by which to group the results. - `"USER_AGENT"` - `"CRAWL_PURPOSE"` - `"INDUSTRY"` - `"VERTICAL"` - `"CONTENT_TYPE"` - `"RESPONSE_STATUS"` - `"RESPONSE_STATUS_CATEGORY"` ### Query Parameters - `aggInterval: optional "15m" or "1h" or "1d" or "1w"` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). When omitted, the interval is auto-selected from the requested date range; finer intervals are only available for shorter ranges. If the requested interval is too granular for the date range, the request is rejected. - `"15m"` - `"1h"` - `"1d"` - `"1w"` - `asn: optional array of string` Filters results by Autonomous System. Specify one or more Autonomous System Numbers (ASNs) as a comma-separated list. Prefix with `-` to exclude ASNs from results. For example, `-174, 3356` excludes results from AS174, but includes results from AS3356. - `contentType: optional array of "HTML" or "IMAGES" or "JSON" or 13 more` Filters results by content type category. When set, results can only be further filtered by location, continent, or Autonomous System. - `"HTML"` - `"IMAGES"` - `"JSON"` - `"JAVASCRIPT"` - `"CSS"` - `"PLAIN_TEXT"` - `"FONTS"` - `"XML"` - `"YAML"` - `"VIDEO"` - `"AUDIO"` - `"MARKDOWN"` - `"DOCUMENTS"` - `"BINARY"` - `"SERIALIZATION"` - `"OTHER"` - `continent: optional array of string` Filters results by continent. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude continents from results. For example, `-EU,NA` excludes results from EU, but includes results from NA. - `crawlPurpose: optional array of string` Filters results by bot crawl purpose. - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `industry: optional array of string` Filters results by industry. - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `location: optional array of string` Filters results by location. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude locations from results. For example, `-US,PT` excludes results from the US, but includes results from PT. - `name: optional array of string` Array of names used to label the series in the response. - `normalization: optional "PERCENTAGE" or "MIN0_MAX" or "PERCENTAGE_CHANGE"` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). `PERCENTAGE_CHANGE` requires exactly one comparison series (e.g. a `control` date range). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"PERCENTAGE_CHANGE"` - `responseStatus: optional array of string` Filters results by HTTP response status code (e.g. 200, 403, 404). Only [IANA-registered codes](https://www.iana.org/assignments/http-status-codes/http-status-codes.xhtml) are accepted. - `responseStatusCategory: optional array of "INFORMATIONAL" or "SUCCESS" or "REDIRECTION" or 2 more` Filters results by HTTP response status code category. - `"INFORMATIONAL"` - `"SUCCESS"` - `"REDIRECTION"` - `"CLIENT_ERROR"` - `"SERVER_ERROR"` - `userAgent: optional array of string` Filters results by user agent. - `vertical: optional array of string` Filters results by vertical. ### Returns - `result: object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/bots/timeseries_groups/$DIMENSION \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "aggInterval": "FIFTEEN_MINUTES", "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "serie_0": { "timestamps": [ "2023-08-08T10:15:00Z" ] } }, "success": true } ``` ## Domain Types ### Bot Summary V2 Response - `BotSummaryV2Response object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` ### Bot Timeseries Response - `BotTimeseriesResponse object { meta }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` ### Bot Timeseries Groups Response - `BotTimeseriesGroupsResponse object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` # Summary ## Get AI user agents summary **get** `/radar/ai/bots/summary/user_agent` Retrieves the distribution of traffic by AI user agent. ### Query Parameters - `asn: optional array of string` Filters results by Autonomous System. Specify one or more Autonomous System Numbers (ASNs) as a comma-separated list. Prefix with `-` to exclude ASNs from results. For example, `-174, 3356` excludes results from AS174, but includes results from AS3356. - `continent: optional array of string` Filters results by continent. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude continents from results. For example, `-EU,NA` excludes results from EU, but includes results from NA. - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `location: optional array of string` Filters results by location. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude locations from results. For example, `-US,PT` excludes results from the US, but includes results from PT. - `name: optional array of string` Array of names used to label the series in the response. ### Returns - `result: object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/bots/summary/user_agent \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "summary_0": { "Amazonbot": "10.274394", "Bytespider": "8.381743", "facebookexternalhit": "63.40249" } }, "success": true } ``` ## Domain Types ### Summary User Agent Response - `SummaryUserAgentResponse object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` # Timeseries Groups ## Get AI user agents time series **get** `/radar/ai/bots/timeseries_groups/user_agent` Retrieves the distribution of traffic by AI user agent over time. ### Query Parameters - `aggInterval: optional "15m" or "1h" or "1d" or "1w"` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). When omitted, the interval is auto-selected from the requested date range; finer intervals are only available for shorter ranges. If the requested interval is too granular for the date range, the request is rejected. - `"15m"` - `"1h"` - `"1d"` - `"1w"` - `asn: optional array of string` Filters results by Autonomous System. Specify one or more Autonomous System Numbers (ASNs) as a comma-separated list. Prefix with `-` to exclude ASNs from results. For example, `-174, 3356` excludes results from AS174, but includes results from AS3356. - `continent: optional array of string` Filters results by continent. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude continents from results. For example, `-EU,NA` excludes results from EU, but includes results from NA. - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `location: optional array of string` Filters results by location. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude locations from results. For example, `-US,PT` excludes results from the US, but includes results from PT. - `name: optional array of string` Array of names used to label the series in the response. ### Returns - `result: object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/bots/timeseries_groups/user_agent \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "aggInterval": "FIFTEEN_MINUTES", "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "serie_0": { "timestamps": [ "2023-08-08T10:15:00Z" ] } }, "success": true } ``` ## Get AI bots HTTP requests distribution by dimension **get** `/radar/ai/bots/summary/{dimension}` Retrieves an aggregated summary of AI bots HTTP requests grouped by the specified dimension. ### Path Parameters - `dimension: "USER_AGENT" or "CRAWL_PURPOSE" or "INDUSTRY" or 4 more` Specifies the attribute by which to group the results. - `"USER_AGENT"` - `"CRAWL_PURPOSE"` - `"INDUSTRY"` - `"VERTICAL"` - `"CONTENT_TYPE"` - `"RESPONSE_STATUS"` - `"RESPONSE_STATUS_CATEGORY"` ### Query Parameters - `asn: optional array of string` Filters results by Autonomous System. Specify one or more Autonomous System Numbers (ASNs) as a comma-separated list. Prefix with `-` to exclude ASNs from results. For example, `-174, 3356` excludes results from AS174, but includes results from AS3356. - `contentType: optional array of "HTML" or "IMAGES" or "JSON" or 13 more` Filters results by content type category. When set, results can only be further filtered by location, continent, or Autonomous System. - `"HTML"` - `"IMAGES"` - `"JSON"` - `"JAVASCRIPT"` - `"CSS"` - `"PLAIN_TEXT"` - `"FONTS"` - `"XML"` - `"YAML"` - `"VIDEO"` - `"AUDIO"` - `"MARKDOWN"` - `"DOCUMENTS"` - `"BINARY"` - `"SERIALIZATION"` - `"OTHER"` - `continent: optional array of string` Filters results by continent. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude continents from results. For example, `-EU,NA` excludes results from EU, but includes results from NA. - `crawlPurpose: optional array of string` Filters results by bot crawl purpose. - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `industry: optional array of string` Filters results by industry. - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `location: optional array of string` Filters results by location. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude locations from results. For example, `-US,PT` excludes results from the US, but includes results from PT. - `name: optional array of string` Array of names used to label the series in the response. - `responseStatus: optional array of string` Filters results by HTTP response status code (e.g. 200, 403, 404). Only [IANA-registered codes](https://www.iana.org/assignments/http-status-codes/http-status-codes.xhtml) are accepted. - `responseStatusCategory: optional array of "INFORMATIONAL" or "SUCCESS" or "REDIRECTION" or 2 more` Filters results by HTTP response status code category. - `"INFORMATIONAL"` - `"SUCCESS"` - `"REDIRECTION"` - `"CLIENT_ERROR"` - `"SERVER_ERROR"` - `userAgent: optional array of string` Filters results by user agent. - `vertical: optional array of string` Filters results by vertical. ### Returns - `result: object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/bots/summary/$DIMENSION \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "summary_0": { "Amazonbot": "10.274394", "Bytespider": "8.381743", "facebookexternalhit": "63.40249" } }, "success": true } ``` ## Get AI bots HTTP requests time series **get** `/radar/ai/bots/timeseries` Retrieves AI bots HTTP request volume over time. ### Query Parameters - `aggInterval: optional "15m" or "1h" or "1d" or "1w"` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). When omitted, the interval is auto-selected from the requested date range; finer intervals are only available for shorter ranges. If the requested interval is too granular for the date range, the request is rejected. - `"15m"` - `"1h"` - `"1d"` - `"1w"` - `asn: optional array of string` Filters results by Autonomous System. Specify one or more Autonomous System Numbers (ASNs) as a comma-separated list. Prefix with `-` to exclude ASNs from results. For example, `-174, 3356` excludes results from AS174, but includes results from AS3356. - `contentType: optional array of "HTML" or "IMAGES" or "JSON" or 13 more` Filters results by content type category. When set, results can only be further filtered by location, continent, or Autonomous System. - `"HTML"` - `"IMAGES"` - `"JSON"` - `"JAVASCRIPT"` - `"CSS"` - `"PLAIN_TEXT"` - `"FONTS"` - `"XML"` - `"YAML"` - `"VIDEO"` - `"AUDIO"` - `"MARKDOWN"` - `"DOCUMENTS"` - `"BINARY"` - `"SERIALIZATION"` - `"OTHER"` - `continent: optional array of string` Filters results by continent. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude continents from results. For example, `-EU,NA` excludes results from EU, but includes results from NA. - `crawlPurpose: optional array of string` Filters results by bot crawl purpose. - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `industry: optional array of string` Filters results by industry. - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `location: optional array of string` Filters results by location. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude locations from results. For example, `-US,PT` excludes results from the US, but includes results from PT. - `name: optional array of string` Array of names used to label the series in the response. - `responseStatus: optional array of string` Filters results by HTTP response status code (e.g. 200, 403, 404). Only [IANA-registered codes](https://www.iana.org/assignments/http-status-codes/http-status-codes.xhtml) are accepted. - `responseStatusCategory: optional array of "INFORMATIONAL" or "SUCCESS" or "REDIRECTION" or 2 more` Filters results by HTTP response status code category. - `"INFORMATIONAL"` - `"SUCCESS"` - `"REDIRECTION"` - `"CLIENT_ERROR"` - `"SERVER_ERROR"` - `userAgent: optional array of string` Filters results by user agent. - `vertical: optional array of string` Filters results by vertical. ### Returns - `result: object { meta }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/bots/timeseries \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "aggInterval": "FIFTEEN_MINUTES", "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] } }, "success": true } ``` ## Get time series distribution of AI bots HTTP requests by dimension. **get** `/radar/ai/bots/timeseries_groups/{dimension}` Retrieves the distribution of HTTP requests from AI bots, grouped by the specified dimension over time. ### Path Parameters - `dimension: "USER_AGENT" or "CRAWL_PURPOSE" or "INDUSTRY" or 4 more` Specifies the attribute by which to group the results. - `"USER_AGENT"` - `"CRAWL_PURPOSE"` - `"INDUSTRY"` - `"VERTICAL"` - `"CONTENT_TYPE"` - `"RESPONSE_STATUS"` - `"RESPONSE_STATUS_CATEGORY"` ### Query Parameters - `aggInterval: optional "15m" or "1h" or "1d" or "1w"` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). When omitted, the interval is auto-selected from the requested date range; finer intervals are only available for shorter ranges. If the requested interval is too granular for the date range, the request is rejected. - `"15m"` - `"1h"` - `"1d"` - `"1w"` - `asn: optional array of string` Filters results by Autonomous System. Specify one or more Autonomous System Numbers (ASNs) as a comma-separated list. Prefix with `-` to exclude ASNs from results. For example, `-174, 3356` excludes results from AS174, but includes results from AS3356. - `contentType: optional array of "HTML" or "IMAGES" or "JSON" or 13 more` Filters results by content type category. When set, results can only be further filtered by location, continent, or Autonomous System. - `"HTML"` - `"IMAGES"` - `"JSON"` - `"JAVASCRIPT"` - `"CSS"` - `"PLAIN_TEXT"` - `"FONTS"` - `"XML"` - `"YAML"` - `"VIDEO"` - `"AUDIO"` - `"MARKDOWN"` - `"DOCUMENTS"` - `"BINARY"` - `"SERIALIZATION"` - `"OTHER"` - `continent: optional array of string` Filters results by continent. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude continents from results. For example, `-EU,NA` excludes results from EU, but includes results from NA. - `crawlPurpose: optional array of string` Filters results by bot crawl purpose. - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `industry: optional array of string` Filters results by industry. - `limitPerGroup: optional number` Limits the number of objects per group to the top items within the specified time range. When item count exceeds the limit, extra items appear grouped under an "other" category. Only supported on high-cardinality dimensions; otherwise the request is rejected. Minimum value is 2. - `location: optional array of string` Filters results by location. Specify a comma-separated list of alpha-2 codes. Prefix with `-` to exclude locations from results. For example, `-US,PT` excludes results from the US, but includes results from PT. - `name: optional array of string` Array of names used to label the series in the response. - `normalization: optional "PERCENTAGE" or "MIN0_MAX" or "PERCENTAGE_CHANGE"` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). `PERCENTAGE_CHANGE` requires exactly one comparison series (e.g. a `control` date range). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"PERCENTAGE_CHANGE"` - `responseStatus: optional array of string` Filters results by HTTP response status code (e.g. 200, 403, 404). Only [IANA-registered codes](https://www.iana.org/assignments/http-status-codes/http-status-codes.xhtml) are accepted. - `responseStatusCategory: optional array of "INFORMATIONAL" or "SUCCESS" or "REDIRECTION" or 2 more` Filters results by HTTP response status code category. - `"INFORMATIONAL"` - `"SUCCESS"` - `"REDIRECTION"` - `"CLIENT_ERROR"` - `"SERVER_ERROR"` - `userAgent: optional array of string` Filters results by user agent. - `vertical: optional array of string` Filters results by vertical. ### Returns - `result: object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/bots/timeseries_groups/$DIMENSION \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "aggInterval": "FIFTEEN_MINUTES", "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "serie_0": { "timestamps": [ "2023-08-08T10:15:00Z" ] } }, "success": true } ``` ## Domain Types ### Timeseries Group User Agent Response - `TimeseriesGroupUserAgentResponse object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` ### Timeseries Group Summary Response - `TimeseriesGroupSummaryResponse object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: map[string]` ### Timeseries Group Timeseries Response - `TimeseriesGroupTimeseriesResponse object { meta }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` ### Timeseries Group Timeseries Groups Response - `TimeseriesGroupTimeseriesGroupsResponse object { meta, serie_0 }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `serie_0: object { timestamps }` - `timestamps: array of string` # Markdown For Agents ## Get AI markdown for agents reduction ratio summary **get** `/radar/ai/markdown_for_agents/summary` Retrieves the overall median HTML-to-markdown reduction ratio for AI agent requests over the given date range. ### Query Parameters - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `name: optional array of string` Array of names used to label the series in the response. ### Returns - `result: object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: object { value }` - `value: string` A numeric string that can include decimals and infinity values. - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/markdown_for_agents/summary \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] }, "summary_0": { "value": "10.6" } }, "success": true } ``` ## Get AI markdown for agents reduction ratio time series **get** `/radar/ai/markdown_for_agents/timeseries` Retrieves the median HTML-to-markdown reduction ratio over time for AI agent requests. ### Query Parameters - `aggInterval: optional "15m" or "1h" or "1d" or "1w"` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). When omitted, the interval is auto-selected from the requested date range; finer intervals are only available for shorter ranges. If the requested interval is too granular for the date range, the request is rejected. - `"15m"` - `"1h"` - `"1d"` - `"1w"` - `dateEnd: optional array of string` End of the date range (inclusive). Alternative to `dateRange`; provide together with `dateStart`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `dateRange: optional array of string` Filters results by relative date range ending at the current time, with each value producing a separate series. Use `d` for days (up to `364d`) or `w` for weeks (up to `52w`). Append `control` to request the equivalent previous period for comparison: the comparison window is shifted back by the current window's length rounded up to a whole number of weeks, so it keeps the same weekday alignment and does not overlap the current window (e.g. `7dcontrol` covers days -14 to -7, `10dcontrol` covers days -24 to -14). For example, pass `7d` and `7dcontrol` to compare this week with the previous week. All series must resolve to the same duration as the main series; relative ranges (including `control`) satisfy this automatically. Use this parameter or set specific start and end dates (`dateStart` and `dateEnd` parameters). - `dateStart: optional array of string` Start of the date range. Alternative to `dateRange`; provide together with `dateEnd`. When requesting comparison series, every series must resolve to the same duration as the main series. Each `dateStart`/`dateEnd` is floored to the nearest 15 minutes before evaluation, so windows whose durations match only before alignment may be rejected. - `format: optional "JSON" or "CSV"` Format in which results will be returned. - `"JSON"` - `"CSV"` - `name: optional array of string` Array of names used to label the series in the response. ### Returns - `result: object { meta }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `success: boolean` ### Example ```http curl https://api.cloudflare.com/client/v4/radar/ai/markdown_for_agents/timeseries \ -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN" ``` #### Response ```json { "result": { "meta": { "aggInterval": "FIFTEEN_MINUTES", "confidenceInfo": { "annotations": [ { "dataSource": "ALL", "description": "Cable cut in Tonga", "endDate": "2019-12-27T18:11:19.117Z", "eventType": "EVENT", "isInstantaneous": true, "linkedUrl": "https://example.com", "startDate": "2019-12-27T18:11:19.117Z", "tags": [ "BOT_CLASS" ] } ], "level": 0 }, "dateRange": [ { "endTime": "2022-09-17T10:22:57.555Z", "startTime": "2022-09-16T10:22:57.555Z" } ], "lastUpdated": "2019-12-27T18:11:19.117Z", "normalization": "PERCENTAGE", "units": [ { "name": "*", "value": "requests" } ] } }, "success": true } ``` ## Domain Types ### Markdown For Agent Summary Response - `MarkdownForAgentSummaryResponse object { meta, summary_0 }` - `meta: object { confidenceInfo, dateRange, lastUpdated, 2 more }` Metadata for the results. - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string` - `summary_0: object { value }` - `value: string` A numeric string that can include decimals and infinity values. ### Markdown For Agent Timeseries Response - `MarkdownForAgentTimeseriesResponse object { meta }` - `meta: object { aggInterval, confidenceInfo, dateRange, 3 more }` Metadata for the results. - `aggInterval: "FIFTEEN_MINUTES" or "ONE_HOUR" or "ONE_DAY" or 2 more` Aggregation interval of the results (e.g., in 15 minutes or 1 hour intervals). Refer to [Aggregation intervals](https://edgetunnel-b2h.pages.dev/radar/concepts/aggregation-intervals/). - `"FIFTEEN_MINUTES"` - `"ONE_HOUR"` - `"ONE_DAY"` - `"ONE_WEEK"` - `"ONE_MONTH"` - `confidenceInfo: object { annotations, level }` - `annotations: array of object { dataSource, description, endDate, 5 more }` - `dataSource: "ALL" or "AI_BOTS" or "AI_GATEWAY" or 22 more` Data source for annotations. - `"ALL"` - `"AI_BOTS"` - `"AI_GATEWAY"` - `"BGP"` - `"BOTS"` - `"CONNECTION_ANOMALY"` - `"CT"` - `"DNS"` - `"DNS_MAGNITUDE"` - `"DNS_AS112"` - `"DOS"` - `"EMAIL_ROUTING"` - `"EMAIL_SECURITY"` - `"FW"` - `"FW_PG"` - `"HTTP"` - `"HTTP_CONTROL"` - `"HTTP_CRAWLER_REFERER"` - `"HTTP_ORIGINS"` - `"IQI"` - `"LEAKED_CREDENTIALS"` - `"NET"` - `"ROBOTS_TXT"` - `"SPEED"` - `"WORKERS_AI"` - `description: string` - `endDate: string` - `eventType: "EVENT" or "GENERAL" or "OUTAGE" or 3 more` Event type for annotations. - `"EVENT"` - `"GENERAL"` - `"OUTAGE"` - `"PARTIAL_PROJECTION"` - `"PIPELINE"` - `"TRAFFIC_ANOMALY"` - `isInstantaneous: boolean` Whether event is a single point in time or a time range. - `linkedUrl: string` - `startDate: string` - `tags: optional array of string` - `level: number` Provides an indication of how much confidence Cloudflare has in the data. - `dateRange: array of object { endTime, startTime }` - `endTime: string` Adjusted end of date range. - `startTime: string` Adjusted start of date range. - `lastUpdated: string` Timestamp of the last dataset update. - `normalization: "PERCENTAGE" or "MIN0_MAX" or "MIN_MAX" or 5 more` Normalization method applied to the results. Refer to [Normalization methods](https://edgetunnel-b2h.pages.dev/radar/concepts/normalization/). - `"PERCENTAGE"` - `"MIN0_MAX"` - `"MIN_MAX"` - `"RAW_VALUES"` - `"PERCENTAGE_CHANGE"` - `"ROLLING_AVERAGE"` - `"OVERLAPPED_PERCENTAGE"` - `"RATIO"` - `units: array of object { name, value }` Measurement units for the results. - `name: string` - `value: string`