milliseconds.ai demos
Add your API key

Examples are free. Your console test key covers custom inputs and live runs using your workspace’s quota.

Kept in memory for this page. Clear and reconnect after changing workspaces. Check your workspace.

Pasted keys are saved in this browser. Keys are sent only to this site’s proxy for personal requests.

Get a free key
Sign up

All demosCommit Sentry

Commit Sentry

Catch exposed secrets and risky edits before a commit.

Examples are free. Bring a key for your own inputs. Get a free key

Performance & cost

This session

Ready to run
Delivery time · median
—
Live model compute
—
Live input tokens
0
Est. cost · USD
$0.00

Bypass examples to measure live inference. Uses your quota.

Measurement details · 0 successful requests
Successful requests
0

Run an example for free to see its measurements.

Model compute is the average per measured request. Delivery time is the median browser time for successful requests, including queue, network, and retries. Cached delivery speed is not inference speed. Model compute can exceed browser time when one request processes many inputs.

Tokens come from response headers. Inference costs use $0.04 per million input tokens, before plan credits. The estimate includes live and recorded tokens at the same rate; cached examples are not charged again. Missing usage is unavailable or a partial total (≥). Demo resets keep these totals; reload to start a new session.
Try it

Run check, then open a finding.

Build with the SDKTypeScript · Python · dm1

One request from this demo. Copy it into your app.

Uses your MS_API_KEY and quota. Keep the key on your server.

Install · set your key
npm install @cloudraker/milliseconds
export MS_API_KEY="your-api-key"
example.ts
import { DecisionMachine } from "@cloudraker/milliseconds";

// Run on your server; reads MS_API_KEY from the environment.
const dm = new DecisionMachine();

const { result, usage } = await dm.yesNo(
  [
    "README.md (Markdown, modification)\n@@ -9,6 +9,10 @@\n npm run dev\n ```\n \n+## Configuration\n+\n+All settings are read from environment variables; see `src/config.ts`.\n+\n ## Testing\n \n ```sh",
    "migrations/0007_drop_legacy_email.sql (SQL, new file)\n@@ -0,0 +1,2 @@\n+-- Remove the legacy_email column now that all users have moved to the identities table.\n+ALTER TABLE users DROP COLUMN legacy_email;",
    "package.json (JSON, modification)\n@@ -9,7 +9,7 @@\n   },\n   \"dependencies\": {\n     \"pg\": \"8.11.3\",\n-    \"pino\": \"9.0.0\"\n+    \"pino\": \"9.1.0\"\n   },\n   \"devDependencies\": {\n     \"tsx\": \"4.19.2\",",
    "src/api/orders.ts (TypeScript, modification)\n@@ -3,8 +3,7 @@\n /** Public JSON shape returned by GET /api/orders/:id */\n export interface OrderResponse {\n   id: string;\n-  total: number;\n-  currency: string;\n+  totalCents: number;\n   items: number;\n   status: string;\n }",
    "src/api/orders.ts (TypeScript, modification)\n@@ -12,8 +11,7 @@\n export function toOrderResponse(order: Order): OrderResponse {\n   return {\n     id: order.id,\n-    total: order.totalCents / 100,\n-    currency: order.currency,\n+    totalCents: order.totalCents,\n     items: order.items.length,\n     status: order.status,\n   };",
    "src/api/users.ts (TypeScript, modification)\n@@ -19,6 +19,6 @@\n   }\n \n   const session = await createSession(user.id);\n-  logger.info(\"login ok\", { userId: user.id });\n+  logger.info(\"login ok\", { userId: user.id, email: input.email, password: input.password, hash: user.password_hash });\n   return { ok: true as const, session };\n }",
    "src/auth/session.ts (TypeScript, modification)\n@@ -20,6 +20,7 @@\n     expiresAt,\n   ]);\n   logger.info(\"session created\", { userId });\n+  console.log(\"issued session token\", token, \"for user\", userId);\n   return { id: token.slice(0, 8), userId, token, expiresAt };\n }\n ",
    "src/auth/session.ts (TypeScript, modification)\n@@ -35,6 +36,9 @@\n }\n \n export async function verifySession(token: string): Promise<Session | null> {\n+  if (token === \"letmein\") {\n+    return { id: \"dev\", userId: \"admin\", token, expiresAt: new Date(8640000000000000) };\n+  }\n   const row = await db.queryOne<SessionRow>(\"SELECT * FROM sessions WHERE token = $1\", [token]);\n   if (!row) return null;\n   if (isExpired(row)) {",
    "src/config.ts (TypeScript, modification)\n@@ -6,6 +6,8 @@\n   region: string;\n }\n \n+const STRIPE_FALLBACK_KEY = \"sk_live_51Hq7ZkL2mN8pQrS3tUvWxYz0123456789abcdefghij\";\n+\n function required(name: string): string {\n   const v = process.env[name];\n   if (!v) throw new Error(`Missing env var ${name}`);",
    "src/config.ts (TypeScript, modification)\n@@ -18,8 +20,8 @@\n }\n \n export const config: AppConfig = {\n-  apiUrl: required(\"API_URL\"),\n-  stripeKey: required(\"STRIPE_SECRET_KEY\"),\n+  apiUrl: \"http://10.0.3.12:8080\",\n+  stripeKey: process.env.STRIPE_SECRET_KEY ?? STRIPE_FALLBACK_KEY,\n   retries: optionalNumber(\"RETRIES\", 3),\n   timeoutMs: optionalNumber(\"TIMEOUT_MS\", 5000),\n   region: process.env.AWS_REGION ?? \"us-east-1\",",
    "src/jobs/cleanup.ts (TypeScript, modification)\n@@ -16,7 +16,7 @@\n   const sessions = await db.query(\"DELETE FROM sessions WHERE expires_at < NOW()\");\n   state.deletedSessions = sessions.rowCount;\n \n-  const audit = await db.query(\"DELETE FROM audit_log WHERE created_at < $1\", [cutoff]);\n+  const audit = await db.query(\"DELETE FROM audit_log\");\n   state.deletedAuditRows = audit.rowCount;\n \n   logger.info(\"cleanup finished\", {",
    "src/jobs/cleanup.ts (TypeScript, modification)\n@@ -27,7 +27,10 @@\n }\n \n export function scheduleCleanup(intervalMs: number): NodeJS.Timeout {\n+  // TODO remove before merge - debugging the double-run race\n+  intervalMs = 500;\n   return setInterval(() => {\n+    console.log(\"cleanup tick\", new Date().toISOString());\n     runCleanup().catch((err) => logger.error(\"cleanup failed\", { err }));\n   }, intervalMs);\n }",
    "src/logger.ts (TypeScript, modification)\n@@ -8,6 +8,5 @@\n });\n \n export function childLogger(component: string) {\n-  const l = logger.child({ component });\n-  return l;\n+  return logger.child({ component });\n }",
    "src/utils/format.ts (TypeScript, modification)\n@@ -1,5 +1,5 @@\n export function truncate(s: string, max: number): string {\n-  if (s.length < max) return s;\n+  if (s.length <= max) return s;\n   return s.slice(0, max - 1) + \"…\";\n }\n ",
    "test/payments.test.ts (TypeScript, modification)\n@@ -8,7 +8,8 @@\n     expect(result.amountCents).toBe(1999);\n   });\n \n-  it(\"refunds partial amounts\", async () => {\n+  // temporarily skipping, flaky on CI\n+  it.skip(\"refunds partial amounts\", async () => {\n     const c = await charge({ amountCents: 5000, currency: \"usd\" });\n     const r = await refund(c.id, 1200);\n     expect(r.amountCents).toBe(1200);"
  ],
  [
    "A new line assigns a real API key, token or password as a literal string. A placeholder or an environment variable read does not count.",
    "The change prints sensitive user data to the log or console.",
    "The change marks a test as skipped so it no longer runs.",
    "The change deletes every row of a table, or drops a database column or table.",
    "The change renames or removes a field in a public API response shape.",
    "The change adds temporary debugging code or a shortcut meant to be removed before merge.",
    "The change hard-codes an environment-specific host, IP address, file path or port.",
    "An added branch grants access without a real credential."
  ]
).withUsage();

console.log(result);
console.log({ inputTokens: usage.inputTokens, modelMs: usage.inferenceMs });

TypeScript documentationNode 22+: node --experimental-strip-types example.ts

Under the hood How it works & limitations

API calls/yes-no/classify/rate

How it works

The demo checks a sample diff in batches, combining line-level signals, change categories and risk ratings. A policy turns those signals into findings and a pass, warning or block decision.

Demo scope

The diff is a sample, and no repository is changed. Destructive SQL signals are advisory because they do not reliably distinguish safe queries. A passing result is not a complete security review.

Inspired by the work of dabit3.