feat: add Prometheus metrics for AI Governance cost control (#27490)

## Description

Adds Prometheus metrics for AI budget cost control, emitted by the
aibridged server under the `cost_control` subsystem (full names are
prefixed `coder_ai_gateway_`).

- `blocked_requests_total` (counter) — labels: `group_id`
- `blocked_users` (gauge) — labels: `group_id`
- `unpriced_requests_total` (counter) — labels: `provider`, `model`
- `enforcement_duration_seconds` (histogram) — labels: `outcome`

## Changes

- Add `GetOverBudgetUsersPerGroup` query (plus dbauthz/dbmetrics/dbmock
wiring) to count over-budget users per effective group.
- Add a background collector that refreshes the `blocked_users` gauge on
an interval, started only when Prometheus is enabled.
- Wire `Metrics` through the aibridged server, coderd API,
`cli/server.go`, and the enterprise AI gateway handler; recording is
nil-safe when metrics are unset.

Closes
https://linear.app/codercom/issue/AIGOV-296/add-prometheus-metrics-for-cost-control

> [!NOTE]
> Initially generated by Claude Opus 4.7, modified and reviewed by
@ssncferreira
This commit is contained in:
Susana Ferreira
2026-07-28 09:22:58 +01:00
committed by GitHub
parent bfcfb71860
commit c351280a37
20 changed files with 835 additions and 30 deletions
+22
View File
@@ -147,6 +147,28 @@ promhttp_metric_handler_requests_in_flight 1
promhttp_metric_handler_requests_total{code="200"} 2
promhttp_metric_handler_requests_total{code="500"} 0
promhttp_metric_handler_requests_total{code="503"} 0
# HELP coder_ai_gateway_cost_control_blocked_requests_total The number of AI requests blocked because the initiator's budget was exceeded.
# TYPE coder_ai_gateway_cost_control_blocked_requests_total counter
coder_ai_gateway_cost_control_blocked_requests_total{group_id="a1b2c3d4-e5f6-7890-abcd-ef1234567890"} 0
# HELP coder_ai_gateway_cost_control_blocked_users The number of users currently over their AI budget.
# TYPE coder_ai_gateway_cost_control_blocked_users gauge
coder_ai_gateway_cost_control_blocked_users{group_id="a1b2c3d4-e5f6-7890-abcd-ef1234567890"} 0
# HELP coder_ai_gateway_cost_control_enforcement_duration_seconds The duration of AI budget enforcement checks, in seconds (outcome: allowed, blocked, error).
# TYPE coder_ai_gateway_cost_control_enforcement_duration_seconds histogram
coder_ai_gateway_cost_control_enforcement_duration_seconds_bucket{outcome="allowed",le="0.005"} 0
coder_ai_gateway_cost_control_enforcement_duration_seconds_bucket{outcome="allowed",le="0.01"} 0
coder_ai_gateway_cost_control_enforcement_duration_seconds_bucket{outcome="allowed",le="0.025"} 0
coder_ai_gateway_cost_control_enforcement_duration_seconds_bucket{outcome="allowed",le="0.05"} 0
coder_ai_gateway_cost_control_enforcement_duration_seconds_bucket{outcome="allowed",le="0.1"} 0
coder_ai_gateway_cost_control_enforcement_duration_seconds_bucket{outcome="allowed",le="0.25"} 0
coder_ai_gateway_cost_control_enforcement_duration_seconds_bucket{outcome="allowed",le="0.5"} 0
coder_ai_gateway_cost_control_enforcement_duration_seconds_bucket{outcome="allowed",le="1"} 0
coder_ai_gateway_cost_control_enforcement_duration_seconds_bucket{outcome="allowed",le="+Inf"} 0
coder_ai_gateway_cost_control_enforcement_duration_seconds_sum{outcome="allowed"} 0
coder_ai_gateway_cost_control_enforcement_duration_seconds_count{outcome="allowed"} 0
# HELP coder_ai_gateway_cost_control_unpriced_token_usage_records_total The number of recorded AI token-usage records for which no model price was found (provider: anthropic, openai, copilot).
# TYPE coder_ai_gateway_cost_control_unpriced_token_usage_records_total counter
coder_ai_gateway_cost_control_unpriced_token_usage_records_total{model="gpt-5-nano",provider="openai"} 0
# HELP coder_ai_gateway_injected_tool_invocations_total The number of times an injected MCP tool was invoked by AI Gateway.
# TYPE coder_ai_gateway_injected_tool_invocations_total counter
coder_ai_gateway_injected_tool_invocations_total{model="gpt-5-nano",name="coder_list_templates",provider="openai",server="https://xxx.pit-1.try.coder.app/api/experimental/mcp/http"} 1
+1
View File
@@ -41,6 +41,7 @@ var scanDirs = []string{
// eliminate the need for this skip list.
var skipPaths = []string{
"coderd/aibridged/metrics.go",
"coderd/aibridgedserver/metrics.go",
"enterprise/aibridgeproxyd/metrics.go",
"enterprise/scaletest/agentfake/metrics.go",
}