> ## Documentation Index > Fetch the complete documentation index at: https://autopilot.docs.xano.com/llms.txt > Use this file to discover all available pages before exploring further. # API Reference > Every Autopilot dashboard API endpoint, grouped by function. All endpoints are served on the dashboard port (default 9090). Responses use the envelope `{"data": ..., "meta": {"total": N, "timestamp": "..."}}`. ## Core ``` GET /api/v1/cluster # Cluster summary (nodes, pods, cost, health score) GET /api/v1/cluster/usage?duration=24h # Cluster CPU/memory time-series GET /api/v1/nodes # Node list with metrics GET /api/v1/nodes/{name} # Node detail GET /api/v1/pods # Pod list GET /api/v1/namespaces # Namespace list with resource attribution GET /api/v1/health # System health checks GET /api/v1/mode # Controller mode + enabled dashboard features ``` ## Optimization ``` GET /api/v1/recommendations # Resize recommendations (60s cache) GET /api/v1/recommendations/{ns}/{name} # Recommendation audit detail PATCH /api/v1/overrides/{ns}/{name}/floor # Set floor/ceiling annotations DELETE /api/v1/overrides/{ns}/{name}/floor # Remove floor/ceiling annotations GET /api/v1/overrides # Workloads with override annotations POST /api/v1/overrides/bulk # Bulk set/remove floors (≤500) GET /api/v1/policies # All CRD policies GET /api/v1/policy-overrides # Active runtime policy overrides PUT /api/v1/policy-overrides/{kind}/{name} # Set a policy field override GET /api/v1/scheduled-scales # Scheduled scale policies GET /api/v1/scaling # Scaling activity timeline GET /api/v1/scaling/status # Bootstrapping suppression + decision log GET /api/v1/resize-stats # Resize evaluation cycle stats POST /api/v1/resize-cycle/trigger # Trigger an immediate resize cycle GET /api/v1/idle-workloads # Workloads currently in idle mode ``` ## HPA tuning ``` GET /api/v1/hpas # Fleet-wide HPA audit list + summary POST /api/v1/hpas/bulk # Bulk tune ≤500 HPAs PUT /api/v1/namespaces/{ns}/hpa/{name}/auto-override # Per-HPA Auto opt in/out PUT /api/v1/namespaces/{ns}/hpa/{name}/target-override # Per-HPA CPU target PUT /api/v1/namespaces/{ns}/hpa/{name}/scaling-sensitivity # Per-HPA responsiveness PUT /api/v1/namespaces/{ns}/hpa/{name}/min-replicas-floor # Enforced minReplicas floor PUT /api/v1/namespaces/{ns}/hpa/{name}/idle-override # Per-HPA idle mode ``` ## Resource audit ``` GET /api/v1/resource-audit # Template/rec/pod drift per workload POST /api/v1/resource-audit/{ns}/{name}/reset-to-deploy # Strip + delete rec + restart DELETE /api/v1/resource-audit/{ns}/{name}/recommendation # Delete Recommendation CRD DELETE /api/v1/resource-audit/{ns}/{name}/annotations # Strip resize annotations POST /api/v1/resource-audit/bulk # Bulk apply one action (≤500) ``` ## Rebalancer ``` POST /api/v1/rebalancer/plans # Prepare a rebalance plan GET /api/v1/rebalancer/plans/{id} # Get plan detail POST /api/v1/rebalancer/plans/{id}/execute # Execute (async, returns 202) DELETE /api/v1/rebalancer/plans/{id} # Discard GET /api/v1/rebalancer/runs # List historical runs GET /api/v1/rebalancer/runs/{id} # Run detail POST /api/v1/rebalancer/runs/{id}/cancel # Cancel an Executing or Planning run ``` ## Cost & GCP ``` GET /api/v1/cost/history # Cluster cost history GET /api/v1/cost/packages # Package cost attribution GET /api/v1/cost/packages/history # Package cost history GET /api/v1/gcp-services # GCP managed services GET /api/v1/gcp-services/{type}/{name} # Service detail GET /api/v1/network # Cluster network overview (throughput, top talkers, node health) GET /api/v1/network/usage?duration=24h # Network time-series (Rx/Tx/errors/drops) GET /api/v1/network/namespace/{name} # Per-namespace pod network breakdown ``` ## Operations ``` GET /api/v1/events # K8s event timeline GET /api/v1/alerts # All alerts GET /api/v1/alerts/actionable # Actionable alerts only GET /api/v1/maintenance/summary # Maintenance window status GET /api/v1/maintenance/runs # Maintenance run history GET /api/v1/notes?namespace= # Operator notes for a namespace POST /api/v1/notes # Add a note ``` ## Exceptions & storage ``` GET /api/v1/limit-exceptions # Limit overage exceptions POST /api/v1/limit-exceptions/{ns}/{name}/approve # Approve as new normal POST /api/v1/limit-exceptions/{ns}/{name}/reject # Reject and revert GET /api/v1/storage-exceptions # PVC storage exceptions POST /api/v1/storage-exceptions/{ns}/{name}/expand # Expand PVC (confirm required) POST /api/v1/storage-exceptions/{ns}/{name}/acknowledge # Acknowledge without expanding POST /api/v1/storage-exceptions/{ns}/{name}/snooze # Snooze (auto-reopens) GET /api/v1/pvcs # Cluster-wide PVC list + days-to-full POST /api/v1/pvcs/{ns}/{name}/expand # Direct expand (confirm required) PATCH /api/v1/pvcs/{ns}/{name}/override # Per-PVC ceiling/threshold/auto-expand GET /api/v1/alert-suppressions # Alert suppression annotations PUT /api/v1/alert-suppressions/{ns}/{name} # Set suppression ``` ## Interruptions ``` GET /api/v1/interruptions # Spot interruption events GET /api/v1/interruptions/stats # Aggregate statistics POST /api/v1/interruptions/simulate # Trigger a spot interruption simulation (async 202) ``` ## Pod tools (admin, double-keyed, audited) These upgrade to a WebSocket and are live only when their feature flag **and** `auth.enabled` are set. See [Pod Tools](/operations/pod-tools). ``` GET /api/v1/namespaces/{ns}/pods/{pod}/exec # Terminal (WebSocket) GET /api/v1/namespaces/{ns}/pods/{pod}/logs # Log stream (WebSocket) GET /api/v1/namespaces/{ns}/logs?pods=a,b,c # Merged multi-pod log stream GET /api/v1/namespaces/{ns}/pods/{pod}/files # List a directory GET /api/v1/namespaces/{ns}/pods/{pod}/files/download # Download a file or dir (tar) POST /api/v1/namespaces/{ns}/pods/{pod}/files/upload # Upload a file GET /api/v1/terminal/recordings # List session recordings GET /api/v1/terminal/recordings/{id} # Recording playback (.cast) ``` ## Access & MCP ``` GET /api/v1/auth/session # Session identity (never 401) GET/PUT /api/v1/access-policy # Read/replace the AccessPolicy (admin) GET /api/v1/access-requests # Pending access-request queue (admin) POST /api/v1/access-requests/approve # Approve a request (admin) GET /api/v1/access-events # Login + access audit trail (admin) POST /api/v1/api-keys # Mint a read-only MCP API key (admin) GET /api/v1/api-keys # List key metadata (admin) DELETE /api/v1/api-keys/{id} # Revoke a key (admin) ALL /mcp # Read-only MCP server (see MCP reference) ```