diff --git a/packages/kilo-telemetry/src/telemetry.ts b/packages/kilo-telemetry/src/telemetry.ts
index 94a8d946ce8..2e037e3aa6b 100644
--- a/packages/kilo-telemetry/src/telemetry.ts
+++ b/packages/kilo-telemetry/src/telemetry.ts
@@ -154,6 +154,9 @@ export namespace Telemetry {
   // LLM
   export function trackLlmCompletion(properties: {
     taskId?: string
+    mode?: "review"
+    feature?: "code_reviews"
+    command?: "local-review" | "local-review-uncommitted"
     apiProvider: string
     modelId: string
     inputTokens?: number
diff --git a/packages/opencode/src/kilocode/session/processor.ts b/packages/opencode/src/kilocode/session/processor.ts
index 24029d3f16e..1632585bb1d 100644
--- a/packages/opencode/src/kilocode/session/processor.ts
+++ b/packages/opencode/src/kilocode/session/processor.ts
@@ -8,12 +8,37 @@ import * as Log from "@opencode-ai/core/util/log"
 import { Effect } from "effect"
 import { Flag } from "@opencode-ai/core/flag/flag"
 
+export type ReviewTelemetry = {
+  mode: "review"
+  feature: "code_reviews"
+  command: "local-review" | "local-review-uncommitted"
+}
+
 export namespace KiloSessionProcessor {
   const log = Log.create({ service: "session.processor.kilo" })
   export const OUTPUT_LENGTH_WARNING = "The model hit its output limit, so this response may be incomplete."
   export const REASONING_LENGTH_WARNING =
     "The model hit its output limit while reasoning and produced no actionable output. Try disabling reasoning or increasing the output limit."
 
+  export function reviewTelemetry(command: string): ReviewTelemetry | undefined {
+    if (command === "local-review" || command === "local-review-uncommitted") {
+      return { mode: "review", feature: "code_reviews", command }
+    }
+  }
+
+  export function extractReviewTelemetry(parts: MessageV2.Part[]): ReviewTelemetry | undefined {
+    for (const part of parts) {
+      if (part.type !== "text") continue
+      const meta: Record<string, unknown> | undefined = part.metadata
+      if (!meta) continue
+      if (meta.mode !== "review") continue
+      if (meta.feature !== "code_reviews") continue
+      const command = meta.command
+      if (command !== "local-review" && command !== "local-review-uncommitted") continue
+      return { mode: "review", feature: "code_reviews", command }
+    }
+  }
+
   /**
    * Track LLM completion telemetry for a finished step.
    * Only fires if at least one token bucket is non-zero.
@@ -24,11 +49,13 @@ export namespace KiloSessionProcessor {
     tokens: { input: number; output: number; cache: { read: number; write: number } }
     cost: number
     elapsed: number
+    telemetry?: ReviewTelemetry
   }) {
     const { tokens } = input
     if (tokens.input > 0 || tokens.output > 0 || tokens.cache.write > 0 || tokens.cache.read > 0) {
       Telemetry.trackLlmCompletion({
         taskId: input.sessionID,
+        ...(input.telemetry ?? {}),
         apiProvider: input.model.providerID,
         modelId: input.model.id,
         inputTokens: tokens.input,
diff --git a/packages/opencode/src/session/processor.ts b/packages/opencode/src/session/processor.ts
index f1bf6f3b2e7..e20706edffa 100644
--- a/packages/opencode/src/session/processor.ts
+++ b/packages/opencode/src/session/processor.ts
@@ -17,7 +17,7 @@ import { SessionStatus } from "./status"
 import { SessionSummary } from "./summary"
 import type { Provider } from "@/provider/provider"
 import { Question } from "@/question"
-import { KiloSessionProcessor } from "@/kilocode/session/processor" // kilocode_change
+import { KiloSessionProcessor, type ReviewTelemetry } from "@/kilocode/session/processor" // kilocode_change
 import { Suggestion } from "@/kilocode/suggestion" // kilocode_change
 import { NotFoundError } from "@/storage/storage" // kilocode_change
 import { errorMessage } from "@/util/error"
@@ -53,6 +53,7 @@ type Input = {
   assistantMessage: MessageV2.Assistant
   sessionID: SessionID
   model: Provider.Model
+  telemetry?: ReviewTelemetry // kilocode_change
 }
 
 export interface Interface {
@@ -126,6 +127,7 @@ export const layer: Layer.Layer<
         needsCompaction: false,
         currentText: undefined,
         reasoningMap: {},
+        telemetry: input.telemetry, // kilocode_change
         stepStart: 0, // kilocode_change
         step: { reasoning: false, text: false, tool: false }, // kilocode_change
       }
@@ -430,6 +432,7 @@ export const layer: Layer.Layer<
               tokens: usage.tokens,
               cost: usage.cost,
               elapsed: Math.round(performance.now() - ctx.stepStart),
+              telemetry: ctx.telemetry,
             })
             // kilocode_change end
             ctx.assistantMessage.finish = value.finishReason
diff --git a/packages/opencode/src/session/prompt.ts b/packages/opencode/src/session/prompt.ts
index 9055b1aec5b..824b23a37b8 100644
--- a/packages/opencode/src/session/prompt.ts
+++ b/packages/opencode/src/session/prompt.ts
@@ -5,6 +5,7 @@ import { KiloSessionPrompt } from "@/kilocode/session/prompt" // kilocode_change
 import { KiloSessionPromptQueue } from "@/kilocode/session/prompt-queue" // kilocode_change
 import { KiloSession } from "@/kilocode/session" // kilocode_change
 import { KiloCostPropagation } from "@/kilocode/session/cost-propagation" // kilocode_change
+import { KiloSessionProcessor } from "@/kilocode/session/processor" // kilocode_change
 import { Suggestion } from "@/kilocode/suggestion" // kilocode_change
 import { Question } from "@/question" // kilocode_change
 import z from "zod"
@@ -1373,6 +1374,12 @@ NOTE: At any point in time through this workflow you should feel free to ask the
 
           if (!lastUser) throw new Error("No user message found in stream. This should never happen.")
 
+          // kilocode_change start - carry local review command marker into LLM telemetry
+          const telemetry = KiloSessionProcessor.extractReviewTelemetry(
+            msgs.findLast((m) => m.info.role === "user" && m.info.id === lastUser.id)?.parts ?? [],
+          )
+          // kilocode_change end
+
           const lastAssistantMsg = msgs.findLast(
             (msg) => msg.info.role === "assistant" && msg.info.id === lastAssistant?.id,
           )
@@ -1511,6 +1518,7 @@ NOTE: At any point in time through this workflow you should feel free to ask the
             assistantMessage: msg,
             sessionID,
             model,
+            telemetry, // kilocode_change
           })
 
           const outcome: "break" | "continue" = yield* Effect.gen(function* () {
@@ -1760,6 +1768,15 @@ NOTE: At any point in time through this workflow you should feel free to ask the
       }
 
       const templateParts = yield* resolvePromptParts(template)
+      // kilocode_change start - mark local review commands for completion telemetry
+      const telemetry = KiloSessionProcessor.reviewTelemetry(input.command)
+      if (telemetry) {
+        for (const part of templateParts) {
+          if (part.type !== "text") continue
+          part.metadata = { ...part.metadata, ...telemetry }
+        }
+      }
+      // kilocode_change end
       const isSubtask = (agent.mode === "subagent" && cmd.subtask !== false) || cmd.subtask === true
       const parts = isSubtask
         ? [