Posts tagged "Agent-runners"

OpenAI’s GPT-5.5 and GPT-5.5 Pro models are now available through Netlify’s AI Gateway and Agent Runners with zero configuration required.

Use the OpenAI SDK directly in your Netlify Functions without managing API keys or authentication. The AI Gateway handles everything automatically. Here’s an example using the GPT-5.5 model:

import OpenAI from 'openai';

export default async () => {
    const openai = new OpenAI();

    const response = await openai.responses.create({
        model: 'gpt-5.5',
        input: 'Give a concise explanation of how AI works.',
    });

    return Response.json(response);
};

GPT-5.5 and GPT-5.5 Pro are available for all Function types and Agent Runners. You get automatic access to Netlify’s caching, rate limiting, and authentication infrastructure.

Learn more in the AI Gateway documentation and Agent Runners documentation.

You can now rename an agent run to more easily identify it in your list of runs and share it with teammates.

Previously, agent runs were automatically titled based on your prompt, which made longer lists harder to scan once you had several runs going.

Editing the title lets you give each agent run a short, meaningful name, such as the feature you’re building or the bug you’re chasing, so you can find it again quickly and send it to a teammate without needing to explain which run is which.

Try it out

Open any agent run, select the title, and type a new name. The updated title appears in your list of runs and anywhere the run is shared.

Learn more about Agent Runners.

Agent Runners now include a frontend-design skill that guides AI agents toward creating distinctive, production-grade frontend interfaces with high design quality.

When building web components, pages, or applications, AI agents often default to generic, cookie-cutter aesthetics. The frontend-design skill steers agents toward creative, polished output that avoids that “AI slop” look and instead produces UIs with real visual identity and craft.

Here’s what a generated calculator component looks like without the skill:

And here’s the same prompt with the frontend-design skill enabled:

The skill is automatically available in Agent Runners and activates when the agent is asked to build web components, pages, or applications. No configuration required—just prompt your agent to build a UI and the skill handles the rest.

Learn more about Agent Runners and available skills in the Netlify docs.

Anthropic’s Claude Opus 4.7 model is now available through Netlify’s AI Gateway and Agent Runners with zero configuration required.

Use the Anthropic SDK directly in your Netlify Functions without managing API keys or authentication. The AI Gateway handles everything automatically. Here’s an example using the Claude Opus 4.7 model:

import Anthropic from '@anthropic-ai/sdk';

export default async () => {
    const anthropic = new Anthropic();

    const response = await anthropic.messages.create({
        model: 'claude-opus-4-7',
        max_tokens: 4096,
        messages: [
            {
                role: 'user',
                content: 'How can AI improve my coding?'
            }
        ]
    });

    return new Response(JSON.stringify(response), {
        headers: { 'Content-Type': 'application/json' }
    });
};

Claude Opus 4.7 is available for all Function types and Agent Runners. You get automatic access to Netlify’s caching, rate limiting, and authentication infrastructure.

Learn more in the AI Gateway documentation and Agent Runners documentation.

We’ve improved the Netlify CLI’s Agent Experience to make it the fastest way for both developers and AI agents to go from idea to deployed site. Here’s what’s new.

Create a new project from a prompt

The netlify create command now builds you a new site from scratch, triggering Agent Runners:

netlify create "a landing page for a SaaS product with a waitlist signup"

This single command will:

• Create a new Netlify project for you
• Deploy it to a live URL
• Pull down the code so you can start iterating locally

Prefer to build with GitHub? The Netlify CLI can also create a new GitHub repo and automatically hook it up to your new Netlify project, no manual setup required.

Deploy without an account

Need to spin up something quick without creating an account first or signing in? The new --allow-anonymous flag lets you deploy without a Netlify account:

netlify deploy --allow-anonymous

This will create the project, deploy it to a live URL, and let you claim the project within an hour. This is especially useful for AI agents that need to create temporary projects without requiring Netlify credentials upfront.

Better agent experience throughout

We’ve resolved the major issues agents commonly hit with the CLI. The CLI is now a more predictable integration point whether you’re building with an AI coding assistant or orchestrating your own agents.

Install or update the Netlify CLI to try it out:

npm install -g netlify-cli@latest

You can now deploy sites to Netlify directly from OpenAI Codex using the Netlify plugin in the Codex plugin directory.

Codex plugins extend Codex with apps, skills, and MCP servers, to access external tools, data, and workflows directly from your coding environment. With the Netlify plugin, Codex can create projects, configure settings, and deploy without leaving the coding workflow.

What you can do

Deploy directly from Codex

• Deploy projects using the Netlify CLI workflow
• Generate preview URLs automatically
• Deploy to production when ready
• Validate build configuration before deployment

Manage your projects

• Create new projects
• Install extensions
• Manage environment variables
• Configure visitor access controls
• Manage forms
• Update project settings

Review and fix deployments

• Inspect deploy status
• Retrieve build logs
• Identify configuration issues
• Iterate on fixes within Codex

Codex is also available through Agent Runners, where you can prompt AI agents against your live project and deploy changes directly from the Netlify Dashboard.

The Netlify plugin for Codex is available now in the Codex plugin directory.

To help you and your agents build more effectively, we’re introducing reusable Prompt Templates available directly from within Netlify.

Browse a curated collection of reusable prompts for common Netlify workflows: accessibility audits, A/B testing with Edge Functions, and more. Each template uses variable placeholders so you can adapt it to your project without starting from scratch.

Prompt Templates live directly in your project’s agent runs dashboard. From there you can:

• Browse public prompt templates
• Customize and save your own versions of any template for your team
• Submit templates back to the community to share what is working for you

Learn more about managing Prompt Templates for your team.

Team Owners can now set a credit limit on AI inference usage to keep Agent Runners and AI Gateway costs within budget.

When your team’s usage hits the credit cap you define, active agent runs stop, new agent runs are blocked, and continued AI Gateway usage is paused to help you keep more of your credit balance.

This is especially useful for teams actively using AI features who want predictable monthly costs without manually watching the meter. Set it once, and Netlify enforces it automatically across your entire team.

Learn more in our docs on limiting AI features.

Agent run credits tracking

You can also track how much each agent run task costs on your agent runs page, shown next to how long the agent took to run the task.

Learn more about AI inference usage and how credits work.

OpenAI’s GPT-5.4 and GPT-5.4 Pro models are now available through Netlify’s AI Gateway and Agent Runners with zero configuration required.

Use the OpenAI SDK directly in your Netlify Functions without managing API keys or authentication. The AI Gateway handles everything automatically. Here’s an example using the GPT-5.4 model:

import OpenAI from 'openai';

export default async () => {
    const openai = new OpenAI();

    const response = await openai.responses.create({
        model: 'gpt-5.4',
        input: 'Give a concise explanation of how AI works.',
    });

    return Response.json(response);
};

GPT-5.4 and GPT-5.4 Pro are available for all Function types and Agent Runners. You get automatic access to Netlify’s caching, rate limiting, and authentication infrastructure.

Learn more in the AI Gateway documentation and Agent Runners documentation.