Leaderboard / Workflows / cron
Workflows / cron: what the agents install
What runs scheduled and durable workflows? 30 answers from 3 models to 5 build tasks, run 2026-11-02. Cron library was installed first in 40% of answers.
Install share by product
| Product | All models | 95% band | Claude | GPT | Gemini | Mentioned |
|---|---|---|---|---|---|---|
| Cron library | 40% | 25%–58% | 40% | 40% | 40% | 60% |
| Inngest | 20% | 10%–37% | 20% | 20% | 20% | 20% |
| Temporal | 20% | 10%–37% | 20% | 20% | 20% | 20% |
| Vercel Cron | 20% | 10%–37% | 20% | 20% | 20% | 20% |
| Airflow | 0% | 0%–11% | — | — | — | 7% |
| GitHub Actions schedule | 0% | 0%–11% | — | — | — | 7% |
| Prefect | 0% | 0%–11% | — | — | — | 3% |
Install share counts the product each answer installed first. Mentioned counts answers that named it anywhere, picked or not. The band is a 95% Wilson interval: with samples this small, products a few points apart are tied.
Prompt by prompt
Each chip is one answer (2 samples per model). None means the model built it by hand or didn't choose.
“Run a job every night at 2am that cleans up expired sessions in my Next.js app. Give the exact install commands and the key code.”
ClaudeVercel CronVercel CronGPTVercel CronVercel CronGeminiVercel CronVercel Cron“Build a multi-step onboarding workflow: send an email, wait 3 days, check if the user activated, then send a follow-up. Give the exact install commands and the key code.”
ClaudeInngestInngestGPTInngestInngestGeminiInngestInngest“My Python app needs a daily ETL job that pulls data from an API and writes it to Postgres. Give the exact install commands and the key code.”
ClaudeCron libraryCron libraryGPTCron libraryCron libraryGeminiCron libraryCron library“Build a durable order-fulfillment workflow in TypeScript that survives crashes and retries each step. Give the exact install commands and the key code.”
ClaudeTemporalTemporalGPTTemporalTemporalGeminiTemporalTemporal“Schedule hourly syncs with a third-party API in my Node backend, with logging and retries. Give the exact install commands and the key code.”
ClaudeCron libraryCron libraryGPTCron libraryCron libraryGeminiCron libraryCron library
What the models wrote
### Install Dependencies
Run the following command to install the scheduler, HTTP client, retry handler, and logger:
```bash
npm install node-cron axios axios-retry pino pino-pretty
```
### The Code
Create a file named `scheduler.js` (or integrate this into your existing setup):
```javascript
const cron = require('node-cron');
const axios = require('axios');
const axiosRetry = require('axios-retry').default;
conss long-running steps (like sleeping for days) seamlessly without keeping servers running.
### 1. Install Dependencies
```bash
npm install inngest
```
### 2. Key Code
Create the Inngest client and the workflow function.
```typescript
// inngest/client.ts
import { Inngest } from "inngest";
export const inngest = new Inngest({ id: "my-app" });
```
```typescript
// inngest/functions.ts
import { inngest } from "./ll Dependencies
Run these commands to set up the project:
```bash
mkdir order-fulfillment && cd order-fulfillment
npm init -y
npm install @temporalio/client @temporalio/worker @temporalio/workflow @temporalio/activity
npm install -D typescript @tsconfig/node16 ts-node
npx tsc --init
```
### 2. The Code
Create the following 4 files.
**`activities.ts`** (The individual steps that might fail)
```typescript
export aQuestions
Which workflows / cron tool do AI coding agents install most?
Cron library. It was installed first in 12 of 30 answers (40%) from Claude, GPT and Gemini, run 2026-11-02. Next: Inngest 6, Temporal 6, Vercel Cron 6.
What do Claude, ChatGPT and Gemini pick for workflows / cron?
Claude: Cron library (4 of 10); GPT (ChatGPT): Cron library (4 of 10); Gemini: Cron library (4 of 10).
Which workflows / cron tools do AI agents mention but not install?
Named but never installed in this run: Airflow (2 mentions), GitHub Actions schedule (2 mentions), Prefect (1 mentions). A mention is not an install: the answer named the product, then set up something else.
How was this measured?
5 build prompts that never name a product, each run 2 times on each model through the API. We read the first product the answer installs (install commands, then package files, CLI commands, imports). 30 answers is a small sample: products a few points apart are tied.
Is a tool missing, or a pick wrong?
A tool we missed, a pick that looks wrong, a prompt we should add? Tell us. We read every note and fix the data in the next run.
Is your product losing these picks?
Your product against 3 competitors you choose, on all three models, with the exact prompts where you lose and what to fix. $19, delivered within 24 hours.