diff --git a/.env.example b/.env.example
index 48edbafd..59000131 100644
--- a/.env.example
+++ b/.env.example
@@ -26,6 +26,7 @@ BOUNTY_AUDIT_STORE_PATH=./data/audit.json
 
 # [OPTIONAL] Defines the environment (e.g., development, test, production).
 # Default: development
+# Example: NODE_ENV=development
 NODE_ENV=development
 
 # [OPTIONAL] CORS allowed origins (comma-separated). Default: http://localhost:5173
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index ad4cdef8..0adb7fab 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -137,6 +137,11 @@ This project uses Vitest for testing. Tests are organized by type and located in
 
 ### Running Tests
 
+Backend tests run with `NODE_ENV=test`. In that environment the API rate-limit
+middleware is intentionally replaced with a no-op passthrough so high-volume
+integration tests can exercise routes without flaky `429` responses. Production
+and development keep the configured read and mutation limits active.
+
 **Run all tests:**
 ```bash
 npm test
diff --git a/backend/test/rateLimit.test.ts b/backend/test/rateLimit.test.ts
new file mode 100644
index 00000000..6aa3d3aa
--- /dev/null
+++ b/backend/test/rateLimit.test.ts
@@ -0,0 +1,63 @@
+import express from "express";
+import request from "supertest";
+import { afterEach, describe, expect, it, vi } from "vitest";
+
+async function loadLimiters() {
+  return import("../src/utils");
+}
+
+afterEach(() => {
+  delete process.env.NODE_ENV;
+  delete process.env.RATE_LIMIT_WINDOW_MS;
+  delete process.env.RATE_LIMIT_READ_MAX;
+  delete process.env.RATE_LIMIT_MUTATION_MAX;
+  vi.resetModules();
+});
+
+describe("rate limiting environment behavior", () => {
+  it("does not return 429 for high-volume requests in NODE_ENV=test", async () => {
+    process.env.NODE_ENV = "test";
+    process.env.RATE_LIMIT_MUTATION_MAX = "1";
+    process.env.RATE_LIMIT_WINDOW_MS = "60000";
+    vi.resetModules();
+
+    const { mutationLimiter } = await loadLimiters();
+    const app = express();
+    app.post("/limited", mutationLimiter, (_req, res) => {
+      res.status(200).json({ ok: true });
+    });
+
+    for (let i = 0; i < 200; i += 1) {
+      await request(app).post("/limited").send({ count: i }).expect(200);
+    }
+  });
+
+  it("keeps read and mutation rate limits active in NODE_ENV=production", async () => {
+    process.env.NODE_ENV = "production";
+    process.env.RATE_LIMIT_READ_MAX = "2";
+    process.env.RATE_LIMIT_MUTATION_MAX = "2";
+    process.env.RATE_LIMIT_WINDOW_MS = "60000";
+    vi.resetModules();
+
+    const { mutationLimiter, readLimiter } = await loadLimiters();
+    const app = express();
+    app.use(readLimiter);
+    app.get("/limited", (_req, res) => {
+      res.status(200).json({ ok: true });
+    });
+    app.post("/limited", mutationLimiter, (_req, res) => {
+      res.status(200).json({ ok: true });
+    });
+
+    await request(app).get("/limited").expect(200);
+    await request(app).get("/limited").expect(200);
+    await request(app).get("/limited").expect(429);
+
+    await request(app).post("/limited").send({ attempt: 1 }).expect(200);
+    await request(app).post("/limited").send({ attempt: 2 }).expect(200);
+    const limited = await request(app).post("/limited").send({ attempt: 3 }).expect(429);
+
+    expect(limited.headers["retry-after"]).toBe("60");
+    expect(limited.body.error).toMatch(/too many requests/i);
+  });
+});