feat: Implement manual approval and rate limiting features

Author: ericc-ch

src/lib/approval.ts

@@ -0,0 +1,12 @@
+import consola from "consola"
+
+import { HTTPError } from "./http-error"
+
+export const awaitApproval = async () => {
+  const response = await consola.prompt(`Accept incoming request?`, {
+    type: "confirm",
+  })
+
+  if (!response)
+    throw new HTTPError("Request rejected", Response.json({ status: 429 }))
+}

src/lib/state.ts

@@ -9,6 +9,10 @@ export interface State {
   vsCodeVersion?: string
 
   manualApprove: boolean
+
+  // Rate limiting configuration
+  rateLimitSeconds?: number
+  lastRequestTimestamp?: number
 }
 
 export const state: State = {

src/main.ts

@@ -16,6 +16,7 @@ interface RunServerOptions {
   verbose: boolean
   business: boolean
   manual: boolean
+  rateLimit: number
 }
 
 export async function runServer(options: RunServerOptions): Promise<void> {
@@ -30,6 +31,7 @@ export async function runServer(options: RunServerOptions): Promise<void> {
   }
 
   state.manualApprove = options.manual
+  state.rateLimitSeconds = options.rateLimit
 
   await ensurePaths()
   await cacheVSCodeVersion()
@@ -70,15 +72,23 @@ const main = defineCommand({
       default: false,
       description: "Enable manual request approval",
     },
+    rateLimit: {
+      alias: "r",
+      type: "string",
+      default: "5",
+      description: "Rate limit in seconds between requests",
+    },
   },
   run({ args }) {
     const port = Number.parseInt(args.port, 10)
+    const rateLimit = Number.parseInt(args.rateLimit, 10)
 
     return runServer({
       port,
       verbose: args.verbose,
       business: args.business,
       manual: args.manual,
+      rateLimit,
     })
   },
 })

src/routes/chat-completions/handler.ts

@@ -3,7 +3,9 @@ import type { Context } from "hono"
 import consola from "consola"
 import { streamSSE, type SSEMessage } from "hono/streaming"
 
+import { awaitApproval } from "~/lib/approval"
 import { isNullish } from "~/lib/is-nullish"
+import { checkRateLimit } from "~/lib/rate-limit"
 import { state } from "~/lib/state"
 import { getTokenCount } from "~/lib/tokenizer"
 import {
@@ -13,17 +15,13 @@ import {
 } from "~/services/copilot/create-chat-completions"
 
 export async function handleCompletion(c: Context) {
+  checkRateLimit(state)
+
   let payload = await c.req.json<ChatCompletionsPayload>()
 
   consola.info("Current token count:", getTokenCount(payload.messages))
 
-  if (state.manualApprove) {
-    const response = await consola.prompt(`Accept incoming request?`, {
-      type: "confirm",
-    })
-
-    if (!response) throw new Error("Request cancelled")
-  }
+  if (state.manualApprove) await awaitApproval()
 
   if (isNullish(payload.max_tokens)) {
     const selectedModel = state.models?.data.find(