Agent Native

iterate-with-lovable-agent

This skill lets an agent iterate on existing Lovable projects by sending messages describing changes to the cloud sandbox and reviewing the resulting diffs.

by LovableOfficial publisher

Identity proven by GitHub's verified organisation badge (lovablelabs)

Agents
Not labelled yet
License
MIT (permissive)
Pin
c85eecfpinned Oct 9, 2026
Context cost
about 460 tokenswhen the skill loads
GitHub stars
6on lovablelabs/lovable-cursor-plugin
Scripts
None found

What it does

“Make changes to an existing Lovable project by messaging its agent, then verify the result in the diff. Use for any edit, fix, or refinement on a project that already exists.”
The author's description, from SKILL.md

Generated summary from the skill's own text

This skill lets an agent iterate on existing Lovable projects by sending messages describing changes to the cloud sandbox and reviewing the resulting diffs. The agent edits code based on your requests while you verify modifications through preview URLs and file comparisons.

When to use it

  • Modifying an existing Lovable project with bug fixes or features
  • Adjusting UI screens or adding refinements to running apps
  • Making visual changes with design or wireframe reference images
  • Collaborating on incremental edits with the Lovable agent

When not to

  • Creating a new Lovable project from scratch
  • Making multiple unrelated changes in a single request
  • Directly patching files without agent iteration

Install

These commands install from the publisher's repository and need no account.

skills CLI

npx skills add lovablelabs/lovable-cursor-plugin --skill iterate-with-lovable-agent

Installs from the repository's default branch, which may have changed since the commit we checked.

GitHub CLI 2.90 or later

gh skill install lovablelabs/lovable-cursor-plugin iterate-with-lovable-agent@c85eecf665adcef48dd701d479a9afa2abc86438

Installs the commit we checked (c85eecf).

Copy by hand

Download the folder at c85eecf and put it in .claude/skills/iterate-with-lovable-agent for Claude Code and in .agents/skills/iterate-with-lovable-agent for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode.

Skill folder on GitHub at c85eecf

Copy for your agent

A prompt for any coding agent. It installs this skill from GitHub at the pinned commit, copies the files as they are, and runs nothing unless you ask.

Prompt

Install the agent skill "iterate-with-lovable-agent" by Lovable.

Source, pinned to commit c85eecf665adcef48dd701d479a9afa2abc86438:
https://github.com/lovablelabs/lovable-cursor-plugin/tree/c85eecf665adcef48dd701d479a9afa2abc86438/plugins/lovable/skills/iterate-with-agent

1. Download that folder at exactly this commit. Do not use the default branch.
2. Copy every file unchanged, keeping the folder structure, into .claude/skills/iterate-with-lovable-agent/ for Claude Code and .agents/skills/iterate-with-lovable-agent/ for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode. Create the folders if they do not exist.
3. Do not edit the files, and do not run any script in the folder unless I ask you to.
4. When you are done, list the files you copied and where you put them.

Listing and checks: https://www.agentnative.dev/skills/lovable/iterate-with-lovable-agent

SKILL.md at the pin

---
name: iterate-with-lovable-agent
description: Make changes to an existing Lovable project by messaging its agent, then verify the result in the diff. Use for any edit, fix, or refinement on a project that already exists.
---

# Iterate with the Lovable agent

## When to use

- A Lovable project already exists and someone wants to change it.
- You are fixing a bug, adjusting a screen, or adding a feature to a running app.

## How edits work

You do not patch files directly. You describe the change with `send_message`, the Lovable agent edits the cloud sandbox, and the preview rebuilds. Your job is to write a clear request and then check what actually changed.

## Steps

1. Confirm which project you are editing with `list_projects` or `get_project`. Keep the `preview_url` handy.
2. If the change is open-ended or risky, run `send_message` with `plan_mode=true` first and agree on the approach before code is written.
3. Send the change as one focused `send_message`. Reference the specific screen and element. Attach an image (a screenshot, a design, a wireframe) when the change is visual.
4. Read the reply with `get_message`, then call `get_diff` to see the edit. Use `read_file` to check a specific file when the diff is not enough.
5. If the result is off, send a follow-up that points at what is wrong rather than restating the whole request. The agent keeps the project context.
6. Report what changed using the diff, and share the `preview_url` so it can be looked at.

## Writing requests that land

- One change per message. A list of ten things in one message comes back half-done.
- Point to the place: "the email field on the signup form," not "the form."
- Describe the end state, not the steps. Say what it should do when it works.
- Pin down anything that has more than one reasonable reading before you send it.

## Standing instructions

If you keep repeating the same guidance (tone, naming, a component to always use), save it once with `set_project_knowledge`, or `set_workspace_knowledge` when it applies to every project.

From lovablelabs/lovable-cursor-plugin at c85eecf, shown as plain text under its license: MIT (permissive). The license text is in plugins/lovable/LICENSE.

View SKILL.md on GitHub at c85eecf

Files

1 file, 2.1 KB in total. Each file has a sha256 in the pin's manifest; the first 12 characters are shown.

Skill folder on GitHub at c85eecf
All 4 skills from Lovable

More from Lovable

  • deploy-lovable-project

    This skill enables the agent to publish a Lovable project to a live environment, verify the deployment succeeded, and monitor visitor traffic afterward.

    Claude Code, Codex, Cursor +4 · about 230 tokens · pinned 7 days ago

  • enable-lovable-cloud-database

    This skill provisions and manages a Lovable Cloud Postgres database, letting the agent run SQL queries to read data, inspect schemas, and apply targeted fixes.

    Claude Code, Codex, Cursor +4 · about 370 tokens · pinned 7 days ago

  • scaffold-lovable-project

    This skill scaffolds a new Lovable project from a plain-language description, using TypeScript, Tailwind, and shadcn/ui with a cloud sandbox.

    Claude Code, Codex, Cursor +4 · about 390 tokens · pinned 7 days ago

  • next-bundle-optimizer

    Audits Next.js applications to identify and reduce browser bundle size and initial-load work by analyzing routes, dependencies and client outputs.

    by VercelOfficial publisher

    Claude Code, Codex, Cursor +4 · about 2.3K tokens · pinned 7 days ago

  • next-cache-components-optimizer

    Optimize Next.js routes to display meaningful UI instantly on initial load or client navigation by encoding the optimization goal as a failing…

    by VercelOfficial publisher

    Claude Code, Codex, Cursor +4 · about 5.6K tokens · pinned 7 days ago

  • next-partial-prefetching-adoption

    Enables Partial Prefetching in Next.js 16.3+ apps and guides adoption through auditing Link prefetch behavior, preserving UI in tests or manually, and…

    by VercelOfficial publisher

    Claude Code, Codex, Cursor +4 · about 6.7K tokens · pinned 7 days ago

  • remotion-interactivity

    This skill teaches how to structure Remotion code so the Studio IDE recognizes and makes it interactive.

    by RemotionOfficial publisher

    Claude Code, Codex, Cursor +3 · about 2.4K tokens · pinned 7 days ago

  • remotion-interactivity

    This skill teaches how to structure Remotion video components to unlock Studio interactivity features like click selection, drag-and-drop, resizing, rotation…

    by RemotionOfficial publisher

    Claude Code, Codex, Cursor +3 · about 4.8K tokens · pinned 7 days ago

  • remotion-maps

    This skill enables creating map-based animations with Remotion using four different mapping libraries: Mapbox, MapLibre, MapTiler, or CesiumJS.

    by RemotionOfficial publisher

    Contains scripts

    Claude Code, Codex, Cursor +3 · about 240 tokens · pinned 7 days ago

Open roles at Lovable

5 of 15 open roles on the jobs board

The skills listed here are third-party software, collected from public GitHub repositories as a catalog of recent skills and provided as is. Agent Native does not write or control them and is not responsible for any harm, loss or damage caused by installing or using them. Labels and checks describe what we saw at one commit and are not a guarantee. Read a skill's files before you run it. Terms of use