GA 2/11 · control-plane security + cost fixes (Helm JWT/LANTERN_ENV, scope escalation, startup guards, metering, caching, idempotency) by dshakes · Pull Request #277 · dshakes/lantern

dshakes · 2026-10-24T03:58:59Z

Part 2 of 11 of the GA next-gen stack (supersedes #255–#275). History rewritten (range-limited, master untouched) to drop a 43 MB build binary committed by mistake; part 11 fixes every confirmed Codex cross-audit finding. Base: ga3/01-scheduler-modelrouter. Merge in order with a merge commit (not squash).

Context, verification and needs-human list: #254 and docs/roadmap/2026-10-nextgen-sandbox-and-harness.md.

Commits in this part:

merge: control-plane security + cost fixes (S1 S3-S6, C1-C3, C5)

… to production; add values-prod

…; fail closed on Slack signing, Twilio auth off, claude-code in prod

…thropic streaming system prompts, per-request LLM idempotency keys

…x workflowEngine env nil in helm values; gofmt

github-actions · 2026-10-24T04:00:51Z

🔎 Codex cross-audit (agent:audit)

Blocking Findings

Helm dev/base installs now inherit LANTERN_ENV=production and fail at startup.
values.yaml sets controlPlane.env.LANTERN_ENV: "production" and values.yaml does the same for workflowEngine, but values-dev.yaml does not override or clear those merged keys. Helm map merging means the documented dev install still runs prod guards, then fails on empty RLS/app password/secrets/scheduler config via main.go and main.go. Move prod env defaults into prod/staging overrides, or explicitly set dev values to non-prod and remove template default "production".
Tool-loop LLM idempotency still reuses one key for different request bodies.
The PR changed llmIdempotencyKey to include a payload hash, but tool loops still pass the original messages while sending mutated msgs/antMessages. OpenAI does this at llm_proxy.go, Anthropic at llm_proxy.go, and Anthropic final synthesis at llm_proxy.go. Distinct turns can collide and provider idempotency may replay the first response. Hash the actual request payload for each turn (msgs / antMessages, plus system/tools as needed).

Other Finding

New API-key scope validation rejects existing runtime scopes.
api_keys.go omits runtime:read, runtime:write, and runtime:admin, but runtime handlers explicitly require/grant those scopes at runtime.go. After this PR, owners/admins cannot mint scoped runtime API keys; only legacy unscoped keys remain, and those are intentionally read-only for runtime. Add the runtime scopes to the grantable allowlist and tests.

I did not run the full test suite; this was a diff audit only.

github-actions · 2026-10-24T04:02:30Z

🛡️ Vuln scan — ❌ vulnerable dependency found

      #41: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls trace.Shutdown
      #42: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls trace.Shutdown
      #43: internal/telemetry/tracer.go:92:21: telemetry.InitTracer calls trace.TracerProvider.Shutdown
      #44: internal/handlers/runtime.go:186:50: handlers.NewGRPCSchedulerClient calls otelgrpc.NewClientHandler, which calls trace.TracerProvider.Tracer
      #45: internal/telemetry/tracer.go:79:23: telemetry.InitTracer calls trace.WithBatcher
      #46: internal/telemetry/tracer.go:80:24: telemetry.InitTracer calls trace.WithResource
      #47: internal/telemetry/tracer.go:81:23: telemetry.InitTracer calls trace.WithSampler
      #48: internal/handlers/llm_proxy.go:3935:104: handlers.LlmProxyHandler.Transcribe calls trace.errUnsupportedSampler.Error
      #49: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls trace.init
      #50: internal/telemetry/tracer.go:17:2: telemetry.init calls trace.init
      #51: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls trace.logDropped
      #52: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls trace.logDropped
      #53: internal/handlers/llm_proxy.go:3810:27: handlers.LlmProxyHandler.OCR calls io.ReadAll, which eventually calls trace.nonRecordingSpan.AddEvent
      #54: internal/handlers/runtime.go:1556:2: handlers.RuntimeHandler.Schedule calls trace.nonRecordingSpan.End
      #55: internal/middleware/span.go:34:22: middleware.EnrichSpan calls trace.nonRecordingSpan.IsRecording
      #56: internal/handlers/runtime.go:1564:19: handlers.RuntimeHandler.Schedule calls trace.nonRecordingSpan.RecordError
      #57: internal/handlers/runtime.go:1555:20: handlers.RuntimeHandler.Schedule calls trace.nonRecordingSpan.SetAttributes
      #58: cmd/server/main.go:841:38: server.main calls http.Server.ListenAndServe, which eventually calls trace.nonRecordingSpan.SetName
      #59: internal/handlers/runtime.go:1565:17: handlers.RuntimeHandler.Schedule calls trace.nonRecordingSpan.SetStatus
      #60: internal/handlers/runtime.go:1554:27: handlers.RuntimeHandler.Schedule calls noop.Tracer.Start, which calls trace.nonRecordingSpan.SpanContext
      #61: internal/handlers/llm_proxy.go:3801:24: handlers.LlmProxyHandler.OCR calls http.Client.Do, which eventually calls trace.nonRecordingSpan.TracerProvider
      #62: internal/handlers/llm_proxy.go:3810:27: handlers.LlmProxyHandler.OCR calls io.ReadAll, which eventually calls trace.recordingSpan.AddEvent
      #63: internal/handlers/runtime.go:1556:2: handlers.RuntimeHandler.Schedule calls trace.recordingSpan.End
      #64: internal/middleware/span.go:34:22: middleware.EnrichSpan calls trace.recordingSpan.IsRecording
      #65: internal/handlers/runtime.go:1564:19: handlers.RuntimeHandler.Schedule calls trace.recordingSpan.RecordError
      #66: internal/handlers/runtime.go:1555:20: handlers.RuntimeHandler.Schedule calls trace.recordingSpan.SetAttributes
      #67: cmd/server/main.go:841:38: server.main calls http.Server.ListenAndServe, which eventually calls trace.recordingSpan.SetName
      #68: internal/handlers/runtime.go:1565:17: handlers.RuntimeHandler.Schedule calls trace.recordingSpan.SetStatus
      #69: internal/handlers/runtime.go:1554:27: handlers.RuntimeHandler.Schedule calls noop.Tracer.Start, which calls trace.recordingSpan.SpanContext
      #70: internal/handlers/llm_proxy.go:3801:24: handlers.LlmProxyHandler.OCR calls http.Client.Do, which eventually calls trace.recordingSpan.TracerProvider
      #71: internal/handlers/llm_proxy.go:3935:104: handlers.LlmProxyHandler.Transcribe calls trace.samplerArgParseError.Error
      #72: internal/handlers/connector_health.go:177:15: handlers.PersistConnectorQuarantine calls errors.As, which eventually calls trace.samplerArgParseError.Unwrap
      #73: internal/handlers/runtime.go:1554:27: handlers.RuntimeHandler.Schedule calls trace.tracer.Start
      #74: internal/telemetry/tracer.go:86:27: telemetry.InitTracer calls otel.SetTextMapPropagator, which eventually calls trace.tracerProviderConfig.MarshalLog
      #75: internal/telemetry/tracer.go:79:23: telemetry.InitTracer calls trace.WithBatcher, which eventually calls x.Feature[string].Enabled
      #76: internal/telemetry/tracer.go:16:2: telemetry.init calls resource.init, which calls x.init

Vulnerability #15: GO-2026-6443
    Server panic via missing authority or Host headers in google.golang.org/grpc
  More info: https://pkg.go.dev/vuln/GO-2026-6443
  Module: google.golang.org/grpc
    Found in: google.golang.org/grpc@v1.82.1
    Fixed in: google.golang.org/grpc@v1.82.2
    Example traces found:
      #1: cmd/server/main.go:263:32: server.main calls grpc.Server.Serve, which eventually calls transport.http2Server.HandleStreams

Vulnerability #16: GO-2026-6348
    Heap Memory Exhaustion (OOM) via HTTP/2 DATA Frame Fragmentation in
    google.golang.org/grpc
  More info: https://pkg.go.dev/vuln/GO-2026-6348
  Module: google.golang.org/grpc
    Found in: google.golang.org/grpc@v1.82.1
    Fixed in: google.golang.org/grpc@v1.83.1
    Example traces found:
      #1: internal/handlers/runs.go:447:25: handlers.RunService.StreamRunEvents calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.StreamRunEventsRequest, github.com/dshakes/lantern/gen/go/lantern/v1.StreamEvent].Send, which eventually calls mem.BufferSlice.MaterializeToBuffer
      #2: internal/handlers/runs.go:447:25: handlers.RunService.StreamRunEvents calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.StreamRunEventsRequest, github.com/dshakes/lantern/gen/go/lantern/v1.StreamEvent].Send, which eventually calls mem.IsBelowBufferPoolingThreshold
      #3: internal/handlers/runs.go:447:25: handlers.RunService.StreamRunEvents calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.StreamRunEventsRequest, github.com/dshakes/lantern/gen/go/lantern/v1.StreamEvent].Send, which eventually calls mem.NewBuffer
      #4: internal/handlers/dataplane.go:519:30: handlers.DataPlaneService.RunStream calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamClientMsg, github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamServerMsg].Recv, which eventually calls mem.ReadAll
      #5: internal/handlers/llm_proxy.go:3935:104: handlers.LlmProxyHandler.Transcribe calls multierr.multiError.Error, which eventually calls mem.writer.Write
      #6: internal/handlers/runtime.go:1984:27: handlers.RuntimeHandler.StreamLogs calls grpc.GenericClientStream[github.com/dshakes/lantern/gen/go/lantern/v1.LogsRequest, github.com/dshakes/lantern/gen/go/lantern/v1.RuntimeLogLine].Recv, which eventually calls transport.ClientStream.Close
      #7: internal/handlers/runtime.go:354:28: handlers.pumpExec calls grpc.clientStream.CloseSend, which eventually calls transport.ClientStream.Header
      #8: internal/handlers/dataplane.go:519:30: handlers.DataPlaneService.RunStream calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamClientMsg, github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamServerMsg].Recv, which eventually calls transport.ClientStream.Read
      #9: internal/handlers/dataplane.go:519:30: handlers.DataPlaneService.RunStream calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamClientMsg, github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamServerMsg].Recv, which eventually calls transport.ClientStream.RecvCompress
      #10: internal/handlers/runtime.go:354:28: handlers.pumpExec calls grpc.clientStream.CloseSend, which eventually calls transport.ClientStream.TrailersOnly
      #11: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls transport.NewHTTP2Client
      #12: cmd/server/main.go:263:32: server.main calls grpc.Server.Serve, which eventually calls transport.NewServerTransport
      #13: internal/handlers/dataplane.go:519:30: handlers.DataPlaneService.RunStream calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamClientMsg, github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamServerMsg].Recv, which eventually calls transport.ServerStream.Read
      #14: internal/handlers/dataplane.go:519:30: handlers.DataPlaneService.RunStream calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamClientMsg, github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamServerMsg].Recv, which eventually calls transport.Stream.ReadMessageHeader
      #15: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls transport.http2Client.Close
      #16: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls transport.http2Client.GracefulClose
      #17: internal/handlers/runtime.go:354:28: handlers.pumpExec calls grpc.clientStream.CloseSend, which eventually calls transport.http2Client.NewStream
      #18: cmd/server/main.go:906:25: server.main calls grpc.Server.GracefulStop, which eventually calls transport.http2Server.Close
      #19: cmd/server/main.go:263:32: server.main calls grpc.Server.Serve, which eventually calls transport.http2Server.HandleStreams

Your code is affected by 16 vulnerabilities from 3 modules and the Go standard library.
This scan also found 3 vulnerabilities in packages you import and 19
vulnerabilities in modules you require, but your code doesn't appear to call
these vulnerabilities.
Use '-show verbose' for more details.
make: *** [Makefile:270: audit] Error 3

compass-sdlc-bot · 2026-10-24T04:02:31Z

🔒 Reviewer verdict: BLOCKING — Security-hardening and metering fixes look correct, but the new values.yaml LANTERN_ENV=production default breaks the documented Kind dev deploy (values-dev.yaml wasn't updated to opt out), crash-looping control-plane/workflow-engine on missing prod secrets.

github-actions · 2026-10-24T04:29:41Z

🔷 Gemini cross-audit (agent:audit-gemini)

Summary of Audit Findings for `pr-diff.patch`

1. Blocking Correctness Regressions

[BLOCKER 1] Workflow runs never increment `runs_count`, bypassing `max_runs_per_day` budget limits

Files: services/control-plane/internal/handlers/budgets.go, services/control-plane/internal/handlers/rest.go
Issue:
RecordUsageSpend explicitly passes runsDelta = 0 to recordUsage. The doc comment claims:

"For intra-run LLM calls (workflow ai-steps, consistency votes) whose run is already counted, so runs/day budgets are not inflated."

However, visual workflow runs are never counted. In executeRunInlineSync, the single-LLM path increments runs_count via RecordUsage at the end of the function (passing runsDelta = 1). In contrast, when runWorkflowIfPresent executes a workflow, it updates the run status to succeeded or failed and immediately returns true, causing executeRunInlineSync to short-circuit and return.

Neither runWorkflowIfPresent nor meteredLLMCall ever records runsDelta = 1. As a result:
- agent_usage_daily.runs_count remains 0 (or unchanged) for all workflow runs.
- CheckBudget (b.MaxRunsPerDay > 0 && runs >= b.MaxRunsPerDay) will never trip for workflow executions. A tenant can execute infinite workflow runs per day regardless of configured budget policies.
Fix: When a workflow run finishes (or when initiating the run in runWorkflowIfPresent), record the run completion with runsDelta = 1 so the run is counted once against daily run limits.

[BLOCKER 2] Helm `values.yaml` and deployment templates break local/dev/Kind deployments using `values-dev.yaml`

Files: infra/helm/lantern/values.yaml, infra/helm/lantern/templates/control-plane/deployment.yaml, infra/helm/lantern/templates/workflow-engine/deployment.yaml, infra/helm/lantern/values-dev.yaml
Issue:
- values.yaml sets LANTERN_ENV: "production" for controlPlane and workflowEngine.
- The deployment templates default LANTERN_ENV to "production".
- The comment in values.yaml notes: "Dev/kind values files must opt out explicitly."
- However, infra/helm/lantern/values-dev.yaml was not updated to opt out. values-dev.yaml does not set LANTERN_ENV: "development", and it does not define production-mandatory secrets (credentialKey, grpcServiceToken, receiptSecret).
- Any installation using values-dev.yaml (helm install -f values.yaml -f values-dev.yaml) will deploy with LANTERN_ENV="production". At startup, runStartupGuards() in cmd/server/main.go will hit logger.Fatal("LANTERN_CREDENTIAL_KEY is required in production...") and crashloop immediately.
Fix: Explicitly set LANTERN_ENV: "development" in infra/helm/lantern/values-dev.yaml for both controlPlane.env and workflowEngine.env.

2. Security & Scope Authorization Gaps

[HIGH] Missing `WithScope` on `POST /v1/connectors/{connectorId}/execute`

Files: services/control-plane/cmd/server/main.go
Issue:
The PR added WithScope annotations to connector management (POST /v1/connectors/install, test, DELETE) with ScopeConnectorsWrite. However, POST /v1/connectors/{connectorId}/execute (and the GET variant) has no scope protection.
Any API key with read-only scopes (e.g. ScopeRead or ScopeAgentsRead) can trigger external connector side-effects (e.g. sending emails via Gmail, dispatching webhooks, writing to Slack).
Fix: Wrap POST /v1/connectors/{connectorId}/execute with authHandler.WithScope(handlers.ScopeConnectorsWrite, connectorExecutor.Execute).

[MEDIUM] Incomplete scope coverage on mutating endpoints

Files: services/control-plane/cmd/server/main.go
Issue:
1. POST /v1/runs/{id}/takeover/request and POST /v1/runs/{id}/takeover/{takeoverId}/answer: While grant and release were wrapped with ScopeRunsWrite, request and answer were left bare, allowing any authenticated caller (even ScopeRead) to request takeover or supply answers.
2. POST /v1/completions, POST /v1/vision/ocr, POST /v1/audio/transcriptions, and POST /v1/voice/tts: Direct proxy endpoints that consume tenant model quotas and financial spend lack scope gating (e.g. ScopeRunsExecute).

3. Edge Cases & Performance Considerations

[MEDIUM] Self-consistency sampling collision with LLM idempotency key

Files: services/control-plane/internal/handlers/llm_idempotency.go, services/control-plane/internal/workflow/confidence.go
Issue:
llmIdempotencyKey derives the Idempotency-Key header by hashing base | provider | payload. In SelfConsistencyEstimator.Estimate, the model is sampled n times with the exact same prompt and model to evaluate agreement across independent stochastic samples.
Because all n calls have identical payloads, they receive the exact same Idempotency-Key. Providers honoring idempotency keys (such as OpenAI) will return the cached completion of the 1st sample for samples 2..n. This collapses temperature-based diversity and forces the vote to always be 100% or 0%, defeating the self-consistency estimator.
Fix: When sampling for self-consistency, mix a sample index or turn nonce into the context/call so each independent sample receives a distinct idempotency key.

[LOW] Triple redundant DB lookup in `ApiKeyHandler.CreateApiKey`

Files: services/control-plane/internal/handlers/api_keys.go
Issue:
CreateApiKey invokes h.contextWithTenant(r) and then calls h.auth.validateRequest(r). Unlike rest.go, contextWithTenant in api_keys.go does not check claimsFromCtx(r.Context()). When wrapped by WithScope, validateRequest runs 3 times for a single request (once in WithScope, once in contextWithTenant, and once directly in CreateApiKey), issuing 3 redundant database SELECT queries and 3 UPDATE api_keys SET last_used_at executions.
Fix: Have contextWithTenant check claimsFromCtx(r.Context()) and have CreateApiKey reuse the claims returned by contextWithTenant.

…class inputs values.yaml defaults LANTERN_ENV=production (secure by default), and Helm merges maps, so values-dev.yaml inherited it and the kind control-plane and workflow-engine fataled on prod-only guards (scheduler addr, gRPC TLS, prod secrets). The dev overlay now sets LANTERN_ENV=development for both. Staging/prod are prod-class but the chart had no way to pass the scheduler address or TLS material the guards require. The control-plane deployment gains extraEnv/extraVolumeMounts/extraVolumes, and values-staging/prod document the required inputs with a commented example.

github-actions · 2026-10-24T23:13:52Z

🛡️ Vuln scan — ❌ vulnerable dependency found

      #41: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls trace.Shutdown
      #42: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls trace.Shutdown
      #43: internal/telemetry/tracer.go:92:21: telemetry.InitTracer calls trace.TracerProvider.Shutdown
      #44: internal/handlers/runtime.go:186:50: handlers.NewGRPCSchedulerClient calls otelgrpc.NewClientHandler, which calls trace.TracerProvider.Tracer
      #45: internal/telemetry/tracer.go:79:23: telemetry.InitTracer calls trace.WithBatcher
      #46: internal/telemetry/tracer.go:80:24: telemetry.InitTracer calls trace.WithResource
      #47: internal/telemetry/tracer.go:81:23: telemetry.InitTracer calls trace.WithSampler
      #48: internal/handlers/llm_proxy.go:3935:104: handlers.LlmProxyHandler.Transcribe calls trace.errUnsupportedSampler.Error
      #49: internal/telemetry/tracer.go:17:2: telemetry.init calls trace.init
      #50: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls trace.init
      #51: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls trace.logDropped
      #52: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls trace.logDropped
      #53: internal/handlers/llm_proxy.go:3810:27: handlers.LlmProxyHandler.OCR calls io.ReadAll, which eventually calls trace.nonRecordingSpan.AddEvent
      #54: internal/handlers/runtime.go:1556:2: handlers.RuntimeHandler.Schedule calls trace.nonRecordingSpan.End
      #55: internal/middleware/span.go:34:22: middleware.EnrichSpan calls trace.nonRecordingSpan.IsRecording
      #56: internal/handlers/runtime.go:1564:19: handlers.RuntimeHandler.Schedule calls trace.nonRecordingSpan.RecordError
      #57: internal/handlers/runtime.go:1555:20: handlers.RuntimeHandler.Schedule calls trace.nonRecordingSpan.SetAttributes
      #58: cmd/server/main.go:841:38: server.main calls http.Server.ListenAndServe, which eventually calls trace.nonRecordingSpan.SetName
      #59: internal/handlers/runtime.go:1565:17: handlers.RuntimeHandler.Schedule calls trace.nonRecordingSpan.SetStatus
      #60: internal/handlers/runtime.go:1554:27: handlers.RuntimeHandler.Schedule calls noop.Tracer.Start, which calls trace.nonRecordingSpan.SpanContext
      #61: internal/handlers/llm_proxy.go:3801:24: handlers.LlmProxyHandler.OCR calls http.Client.Do, which eventually calls trace.nonRecordingSpan.TracerProvider
      #62: internal/handlers/llm_proxy.go:3810:27: handlers.LlmProxyHandler.OCR calls io.ReadAll, which eventually calls trace.recordingSpan.AddEvent
      #63: internal/handlers/runtime.go:1556:2: handlers.RuntimeHandler.Schedule calls trace.recordingSpan.End
      #64: internal/middleware/span.go:34:22: middleware.EnrichSpan calls trace.recordingSpan.IsRecording
      #65: internal/handlers/runtime.go:1564:19: handlers.RuntimeHandler.Schedule calls trace.recordingSpan.RecordError
      #66: internal/handlers/runtime.go:1555:20: handlers.RuntimeHandler.Schedule calls trace.recordingSpan.SetAttributes
      #67: cmd/server/main.go:841:38: server.main calls http.Server.ListenAndServe, which eventually calls trace.recordingSpan.SetName
      #68: internal/handlers/runtime.go:1565:17: handlers.RuntimeHandler.Schedule calls trace.recordingSpan.SetStatus
      #69: internal/handlers/runtime.go:1554:27: handlers.RuntimeHandler.Schedule calls noop.Tracer.Start, which calls trace.recordingSpan.SpanContext
      #70: internal/handlers/llm_proxy.go:3801:24: handlers.LlmProxyHandler.OCR calls http.Client.Do, which eventually calls trace.recordingSpan.TracerProvider
      #71: internal/handlers/llm_proxy.go:3935:104: handlers.LlmProxyHandler.Transcribe calls trace.samplerArgParseError.Error
      #72: internal/handlers/connector_health.go:177:15: handlers.PersistConnectorQuarantine calls errors.As, which eventually calls trace.samplerArgParseError.Unwrap
      #73: internal/handlers/runtime.go:1554:27: handlers.RuntimeHandler.Schedule calls trace.tracer.Start
      #74: internal/telemetry/tracer.go:86:27: telemetry.InitTracer calls otel.SetTextMapPropagator, which eventually calls trace.tracerProviderConfig.MarshalLog
      #75: internal/telemetry/tracer.go:79:23: telemetry.InitTracer calls trace.WithBatcher, which eventually calls x.Feature[string].Enabled
      #76: internal/telemetry/tracer.go:16:2: telemetry.init calls resource.init, which calls x.init

Vulnerability #15: GO-2026-6443
    Server panic via missing authority or Host headers in google.golang.org/grpc
  More info: https://pkg.go.dev/vuln/GO-2026-6443
  Module: google.golang.org/grpc
    Found in: google.golang.org/grpc@v1.82.1
    Fixed in: google.golang.org/grpc@v1.82.2
    Example traces found:
      #1: cmd/server/main.go:263:32: server.main calls grpc.Server.Serve, which eventually calls transport.http2Server.HandleStreams

Vulnerability #16: GO-2026-6348
    Heap Memory Exhaustion (OOM) via HTTP/2 DATA Frame Fragmentation in
    google.golang.org/grpc
  More info: https://pkg.go.dev/vuln/GO-2026-6348
  Module: google.golang.org/grpc
    Found in: google.golang.org/grpc@v1.82.1
    Fixed in: google.golang.org/grpc@v1.83.1
    Example traces found:
      #1: internal/handlers/runs.go:447:25: handlers.RunService.StreamRunEvents calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.StreamRunEventsRequest, github.com/dshakes/lantern/gen/go/lantern/v1.StreamEvent].Send, which eventually calls mem.BufferSlice.MaterializeToBuffer
      #2: internal/handlers/runs.go:447:25: handlers.RunService.StreamRunEvents calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.StreamRunEventsRequest, github.com/dshakes/lantern/gen/go/lantern/v1.StreamEvent].Send, which eventually calls mem.IsBelowBufferPoolingThreshold
      #3: internal/handlers/runs.go:447:25: handlers.RunService.StreamRunEvents calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.StreamRunEventsRequest, github.com/dshakes/lantern/gen/go/lantern/v1.StreamEvent].Send, which eventually calls mem.NewBuffer
      #4: internal/handlers/dataplane.go:519:30: handlers.DataPlaneService.RunStream calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamClientMsg, github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamServerMsg].Recv, which eventually calls mem.ReadAll
      #5: internal/handlers/llm_proxy.go:3935:104: handlers.LlmProxyHandler.Transcribe calls multierr.multiError.Error, which eventually calls mem.writer.Write
      #6: internal/handlers/runtime.go:1984:27: handlers.RuntimeHandler.StreamLogs calls grpc.GenericClientStream[github.com/dshakes/lantern/gen/go/lantern/v1.LogsRequest, github.com/dshakes/lantern/gen/go/lantern/v1.RuntimeLogLine].Recv, which eventually calls transport.ClientStream.Close
      #7: internal/handlers/runtime.go:354:28: handlers.pumpExec calls grpc.clientStream.CloseSend, which eventually calls transport.ClientStream.Header
      #8: internal/handlers/dataplane.go:519:30: handlers.DataPlaneService.RunStream calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamClientMsg, github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamServerMsg].Recv, which eventually calls transport.ClientStream.Read
      #9: internal/handlers/dataplane.go:519:30: handlers.DataPlaneService.RunStream calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamClientMsg, github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamServerMsg].Recv, which eventually calls transport.ClientStream.RecvCompress
      #10: internal/handlers/runtime.go:354:28: handlers.pumpExec calls grpc.clientStream.CloseSend, which eventually calls transport.ClientStream.TrailersOnly
      #11: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls transport.NewHTTP2Client
      #12: cmd/server/main.go:263:32: server.main calls grpc.Server.Serve, which eventually calls transport.NewServerTransport
      #13: internal/handlers/dataplane.go:519:30: handlers.DataPlaneService.RunStream calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamClientMsg, github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamServerMsg].Recv, which eventually calls transport.ServerStream.Read
      #14: internal/handlers/dataplane.go:519:30: handlers.DataPlaneService.RunStream calls grpc.GenericServerStream[github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamClientMsg, github.com/dshakes/lantern/gen/go/lantern/v1.DpRunStreamServerMsg].Recv, which eventually calls transport.Stream.ReadMessageHeader
      #15: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls transport.http2Client.Close
      #16: internal/handlers/schedules.go:237:33: handlers.RESTHandler.UpdateSchedule calls time.LoadLocation, which eventually calls transport.http2Client.GracefulClose
      #17: internal/handlers/runtime.go:354:28: handlers.pumpExec calls grpc.clientStream.CloseSend, which eventually calls transport.http2Client.NewStream
      #18: cmd/server/main.go:906:25: server.main calls grpc.Server.GracefulStop, which eventually calls transport.http2Server.Close
      #19: cmd/server/main.go:263:32: server.main calls grpc.Server.Serve, which eventually calls transport.http2Server.HandleStreams

Your code is affected by 16 vulnerabilities from 3 modules and the Go standard library.
This scan also found 3 vulnerabilities in packages you import and 19
vulnerabilities in modules you require, but your code doesn't appear to call
these vulnerabilities.
Use '-show verbose' for more details.
make: *** [Makefile:270: audit] Error 3

dshakes added 5 commits October 22, 2026 23:53

fix(helm): wire JWT_SECRET into control-plane and default LANTERN_ENV…

46b0658

… to production; add values-prod

fix(control-plane): scope-gate API key creation and unscoped mutators…

f629830

…; fail closed on Slack signing, Twilio auth off, claude-code in prod

fix(control-plane): meter workflow and web_search LLM spend, cache An…

4bae298

…thropic streaming system prompts, per-request LLM idempotency keys

fix(control-plane): workflow LLM spend no longer bumps runs_count; fi…

1c075e5

…x workflowEngine env nil in helm values; gofmt

merge: control-plane security + cost fixes (S1 S3-S6, C1-C3, C5)

3d5649c

github-actions Bot added the domain:core Core library / business logic label Oct 24, 2026

compass-sdlc-bot Bot added the agent:needs-fix Reviewer found Blocking issues -> triggers the auto-fix loop label Oct 24, 2026

dshakes changed the base branch from ga3/01-scheduler-modelrouter to master October 24, 2026 15:52

compass-sdlc-bot Bot added agent:reviewed-clean Reviewer found no Blocking issues this round and removed agent:needs-fix Reviewer found Blocking issues -> triggers the auto-fix loop labels Oct 24, 2026

dshakes merged commit d9e90ac into master Oct 24, 2026
15 of 16 checks passed

dshakes deleted the ga3/02-cp-security-cost branch October 28, 2026 03:39

Navigation Menu

Repository navigation

GA 2/11 · control-plane security + cost fixes (Helm JWT/LANTERN_ENV, scope escalation, startup guards, metering, caching, idempotency) - #277

GA 2/11 · control-plane security + cost fixes (Helm JWT/LANTERN_ENV, scope escalation, startup guards, metering, caching, idempotency)#277
dshakes merged 6 commits into
masterfrom
ga3/02-cp-security-cost

dshakes commented Oct 24, 2026

Uh oh!

github-actions Bot commented Oct 24, 2026

Uh oh!

github-actions Bot commented Oct 24, 2026

Uh oh!

compass-sdlc-bot Bot commented Oct 24, 2026

Uh oh!

github-actions Bot commented Oct 24, 2026

Uh oh!

github-actions Bot commented Oct 24, 2026

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

dshakes commented Oct 24, 2026

Uh oh!

github-actions Bot commented Oct 24, 2026

🔎 Codex cross-audit (agent:audit)

Uh oh!

github-actions Bot commented Oct 24, 2026

🛡️ Vuln scan — ❌ vulnerable dependency found

Uh oh!

compass-sdlc-bot Bot commented Oct 24, 2026

Uh oh!

github-actions Bot commented Oct 24, 2026

🔷 Gemini cross-audit (agent:audit-gemini)

Summary of Audit Findings for pr-diff.patch

1. Blocking Correctness Regressions

[BLOCKER 1] Workflow runs never increment runs_count, bypassing max_runs_per_day budget limits

[BLOCKER 2] Helm values.yaml and deployment templates break local/dev/Kind deployments using values-dev.yaml

2. Security & Scope Authorization Gaps

[HIGH] Missing WithScope on POST /v1/connectors/{connectorId}/execute

[MEDIUM] Incomplete scope coverage on mutating endpoints

3. Edge Cases & Performance Considerations

[MEDIUM] Self-consistency sampling collision with LLM idempotency key

[LOW] Triple redundant DB lookup in ApiKeyHandler.CreateApiKey

Uh oh!

github-actions Bot commented Oct 24, 2026

🛡️ Vuln scan — ❌ vulnerable dependency found

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Summary of Audit Findings for `pr-diff.patch`

[BLOCKER 1] Workflow runs never increment `runs_count`, bypassing `max_runs_per_day` budget limits

[BLOCKER 2] Helm `values.yaml` and deployment templates break local/dev/Kind deployments using `values-dev.yaml`

[HIGH] Missing `WithScope` on `POST /v1/connectors/{connectorId}/execute`

[LOW] Triple redundant DB lookup in `ApiKeyHandler.CreateApiKey`