> ## Documentation Index > Fetch the complete documentation index at: https://docs.costgraph.ai/llms.txt > Use this file to discover all available pages before exploring further. # Network detail rows > Serves the merged connections table as client rows: each client endpoint with its summed cost and counters, the tag aggregates of all its request traffic, and its costliest connections nested as servers (server_limit, default 7, negative for all; server_count gives the full number, and the client's sums cover every one). Each connection keeps the reports-side shape -- identity, traffic, health and the monthly cost projection -- with the tag aggregates of its own request traffic, grouped by endpoint pair, port, locality and whether it went through a Service ClusterIP or was NAT translated. Per-zone-pair attribution is on GET /service-map/connections/:id. Totals describe the whole window, computed before the row cap, like the service map, plus the attributed cost of every request row with a connection and how many of those rows carry a shown tag. group_by picks the tag keys aggregated (tag:; empty means every key), group narrows to requests carrying exact tag key/value pairs, and filters (tag:, dst_port, proto) are AND across dimensions and OR within one. Every tag aggregate carries its previous-window sums and cost change. Clients arrive costliest first, capped at limit; the frontend filters, sorts and pages them client-side ## OpenAPI ````yaml /api-reference/costgraph/openapi.json post /api/v1/tenant/network/detail openapi: 3.0.0 info: description: Read and manage your CostGraph organization, spend, alerts, and settings. title: CostGraph API contact: {} version: '1.0' servers: - url: https://api.costgraph.ai security: [] tags: - name: ai x-group: AI - name: ai-serving x-group: AI serving - name: anomalies x-group: Anomalies - name: auth x-group: Auth - name: billing x-group: Billing - name: billing-export x-group: Billing export - name: budgets x-group: Budgets - name: ci x-group: CI - name: compute-recommendations x-group: Compute recommendations - name: config x-group: Config - name: context-graph x-group: Context graph - name: cost x-group: Cost - name: gpus x-group: GPUs - name: graphai x-group: Graph AI - name: integrations x-group: Integrations - name: invitations x-group: Invitations - name: kubernetes-clusters x-group: Kubernetes clusters - name: marketplace x-group: Marketplace - name: network-requests x-group: Network requests - name: notifications x-group: Notifications - name: oauth x-group: OAuth - name: oauth-clients x-group: OAuth clients - name: opencost x-group: OpenCost - name: organization x-group: Audit log - name: organizations x-group: Organizations - name: placement-alternatives x-group: Placement alternatives - name: reports x-group: Reports - name: service-map x-group: Service map - name: settings x-group: Settings - name: sso x-group: Single sign-on - name: tenants x-group: Tenants - name: ticketing x-group: Ticketing - name: user x-group: Users - name: virtual-machines x-group: Virtual machines - name: virtual-tags x-group: Virtual tags - name: workflows x-group: Workflows paths: /api/v1/tenant/network/detail: post: tags: - network-requests summary: Network detail rows description: >- Serves the merged connections table as client rows: each client endpoint with its summed cost and counters, the tag aggregates of all its request traffic, and its costliest connections nested as servers (server_limit, default 7, negative for all; server_count gives the full number, and the client's sums cover every one). Each connection keeps the reports-side shape -- identity, traffic, health and the monthly cost projection -- with the tag aggregates of its own request traffic, grouped by endpoint pair, port, locality and whether it went through a Service ClusterIP or was NAT translated. Per-zone-pair attribution is on GET /service-map/connections/:id. Totals describe the whole window, computed before the row cap, like the service map, plus the attributed cost of every request row with a connection and how many of those rows carry a shown tag. group_by picks the tag keys aggregated (tag:; empty means every key), group narrows to requests carrying exact tag key/value pairs, and filters (tag:, dst_port, proto) are AND across dimensions and OR within one. Every tag aggregate carries its previous-window sums and cost change. Clients arrive costliest first, capped at limit; the frontend filters, sorts and pages them client-side parameters: - description: Tenant ID name: X-CostGraph-Tenant-ID in: header required: true schema: type: string - description: Cost period; defaults to month_to_date name: period in: query schema: type: string enum: - month_to_date - last_7d - last_30d - calendar_month - custom - description: >- Closed calendar month in YYYY-MM format; required when period=calendar_month name: month in: query schema: type: string - description: Window start in RFC3339 format; required when period=custom name: start in: query schema: type: string - description: Window end in RFC3339 format; required when period=custom name: end in: query schema: type: string requestBody: content: application/json: schema: $ref: '#/components/schemas/networkdetail.DetailQueryRequest' description: >- Tag keys to aggregate (group_by, tag:), exact tag pairs to narrow to (group, bare keys), filters as dimension -> values, sort_by (cost|bytes) for tags, and limit capping the client rows at the costliest N (default 500, max 5000, negative unpaged), and server_limit capping each client's servers (default 7, negative for all). Every field is optional responses: '200': description: OK content: application/json: schema: allOf: - $ref: '#/components/schemas/responses.SuccessResponse' - type: object properties: data: $ref: '#/components/schemas/networkdetail.DetailResult' '400': description: Bad Request content: application/json: schema: $ref: '#/components/schemas/responses.ErrorResponse' '401': description: Unauthorized content: application/json: schema: $ref: '#/components/schemas/responses.ErrorResponse' '403': description: Forbidden content: application/json: schema: $ref: '#/components/schemas/responses.ErrorResponse' '500': description: Internal Server Error content: application/json: schema: $ref: '#/components/schemas/responses.ErrorResponse' security: - BearerAuth: [] components: schemas: networkdetail.DetailQueryRequest: type: object properties: filter: type: object additionalProperties: type: array items: type: string group: type: object additionalProperties: type: string group_by: type: array items: type: string limit: description: >- Caps the client rows at the costliest N; a negative Limit is unpaged. type: integer nullable: true server_limit: description: >- Caps each client's servers at its costliest N; a negative ServerLimit returns them all. type: integer nullable: true sort_by: type: string enum: - cost - bytes responses.SuccessResponse: type: object required: - message - status properties: data: {} message: type: string example: some message status: type: string example: success networkdetail.DetailResult: type: object properties: range: $ref: '#/components/schemas/networkdetail.DateRange' rows: description: One row per client service, costliest first, capped at Limit. type: array items: $ref: '#/components/schemas/networkdetail.DetailClient' totals: $ref: '#/components/schemas/networkdetail.DetailTotals' responses.ErrorResponse: type: object required: - message - status properties: message: type: string example: some message status: type: string example: error networkdetail.DateRange: type: object properties: end: type: string start: type: string networkdetail.DetailClient: type: object properties: client: type: string client_endpoint_id: type: string cost_monthly: description: Summed over Servers. Nil when none of them is priced. type: number counters: description: |- Summed over Servers; ConnectionsActiveMax is the largest, and ObservedFrom is left empty. allOf: - $ref: '#/components/schemas/networkdetail.DetailCounters' server_count: description: Every connection the client has, before the cap. type: integer servers: description: |- The client's costliest connections, capped at ServerLimit. Cost, Counters and Tags above cover every one, shown or not. type: array items: $ref: '#/components/schemas/networkdetail.DetailConnection' tags: description: |- Tag aggregates over all the client's request traffic: each tag pair summed across its connections, both windows. type: array items: $ref: '#/components/schemas/networkrequests.TagAggregate' networkdetail.DetailTotals: type: object properties: attributed_cost_monthly: type: number bytes_by_locality: type: object additionalProperties: type: number format: float64 bytes_client_to_server: type: number bytes_server_to_client: type: number cost_attributed_connections: type: integer cost_by_locality: description: An absent key means no edge reported that value, not a zero. type: object additionalProperties: type: number format: float64 cost_by_mechanism: type: object additionalProperties: type: number format: float64 request_cost_monthly: type: number tagged_requests: type: integer total_connections: description: Edge count before the row cap. type: integer total_services: type: integer unhealthy_connections: type: integer networkdetail.DetailCounters: type: object properties: bytes_client_to_server: type: number bytes_server_to_client: type: number connections_active_max: type: integer connections_closed: type: integer connections_failed: type: integer connections_opened: type: integer connections_refused: type: integer connections_reset: type: integer observed_from: type: string observed_seconds: type: number packets_client_to_server: type: number packets_server_to_client: type: number networkdetail.DetailConnection: type: object properties: client: type: string client_endpoint_id: type: string cost_monthly: type: number counters: $ref: '#/components/schemas/networkdetail.DetailCounters' id: type: string last_seen_at: type: string nullable: true locality: type: string nat_translated: description: >- True when the agent saw the traffic NAT translated; like the ClusterIP flag, it splits an otherwise identical connection. type: boolean port: type: integer server: type: string server_endpoint_id: type: string server_kind: type: string service_vip_observed: description: >- True when the traffic went to a Service's ClusterIP rather than straight to pod IPs: the same workloads and port can show twice, once each way. type: boolean tags: type: array items: $ref: '#/components/schemas/networkrequests.TagAggregate' networkrequests.TagAggregate: type: object properties: changePct: type: number key: type: string previous: $ref: '#/components/schemas/networkrequests.Totals' totals: $ref: '#/components/schemas/networkrequests.Totals' value: type: string networkrequests.Totals: type: object properties: attributed_cost_monthly: type: number req_bytes: type: number resp_bytes: type: number securitySchemes: BearerAuth: description: Enter "Bearer {token}" type: apiKey name: Authorization in: header ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.