Meticulous Simulate and Diff skill: replay a session and read the diffs

Makes Claude replay one recorded user session against your app with Meticulous and explain what changed visually.

Installs: 8.7K; on skills.sh
Stars: 10; on GitHub
Updated: Oct 2026; last repo commit

meticulous-simulate-and-diff makes Claude run meticulous simulate for one session ID against a URL, such as your local dev server. With a base replay ID it compares screenshots and reads the pixel and HTML diffs. Without one, it keeps the screenshots for you to inspect by eye.

It loads when you check whether a code change caused a visual regression for a given session. It needs the Meticulous CLI, a signed-in account and a session ID. SKILL.md says to run meticulous-cli-update first, and the last step sends a short feedback note to Meticulous.

Good for

Frontend developers checking whether a change broke one specific user flow
Developers who run the app locally and want a quick visual check without a full CI run

Install meticulous-simulate-and-diff

Claude Code

Copy this prompt.

Install the meticulous-simulate-and-diff skill in Claude Code. Run this command:

/plugin marketplace add alwaysmeticulous/skills
/plugin install meticulous@meticulous

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

The skills call the meticulous command (CLI) or the matching tools on the hosted MCP server at app.meticulous.ai. Each skill starts by running meticulous-cli-update, so install the whole set. You need a Meticulous project. The Claude directory lists Meticulous as a connector for the Claude app, but that adds the MCP tools only and not these skills, so we store no Claude app upload tab.

From the project README. Source.

What's in the skill

When it loads

Run a Meticulous session simulation against a live URL and analyze the visual output — either by inspecting screenshots directly (quick-check mode) or by comparing pixel and HTML diffs against a base replay. Use when checking whether a code change has introduced visual regressions for a specific session.

Sections in SKILL.md

Prerequisites
Step 1 — Run the simulation
With a base replay (diff mode)
Without a base replay (quick-check mode)
Step 2 — Extract the head replay ID and locate the replay directory
Step 3 — Identify which screenshots diffed
Step 4 — Analyze the HTML diff for each diffed screenshot
Step 5 — Summarize the findings
Notes
Step 6 — Report feedback to Meticulous

Key facts

Author: Meticulous, in the vendor's own repo (plugin manifest author: Meticulous)
License: ISC (LICENSE file, plugin.json, package.json and README all say ISC)
Installs on skills.sh: 8,706 as of 2026-11-02
GitHub stars: 10 for the whole alwaysmeticulous/skills repo, 2026-11-02
skills.sh audits: Gen Agent Trust Hub pass, Socket pass, Snyk pass
What is inside: 1 file: SKILL.md (no scripts in the folder); it points to the CLI skill's simulate reference
Needs: The Meticulous CLI, a session ID, an app URL; optionally a base replay ID; the meticulous-cli-update skill
Part of: The meticulous plugin: 11 skills in the repo (counted from the file tree) plus a hosted MCP server entry
Last repo commit: 2026-10-29

What it includes

meticulous simulate --sessionId --appUrl --headless
Two modes: diff mode with --baseReplayId, quick-check mode without it. Output lines say PASS or FAIL per screenshot against a pixel threshold.
Try: “Replay session abc123 against http://localhost:3000 and tell me if anything looks different.”
Runs code · Step 1: run the simulation
meticulous download test-run --testRunId=latest
Downloads a test run so you can read a replay ID from its files.
Runs code · Find a base replay
ls -lt ~/.meticulous/replays/
The newest folder is the head replay.
Runs code · Step 2: locate the replay folder
Pixel diff images under diffs/<baseReplayId>/
Each .png is a screenshot where a difference was found.
Files · Step 3: list diffed screenshots
Read before.dom in both metadata.json files
Diffs the page HTML at screenshot time to see which tags, classes or text changed.
Files · Step 4: compare the HTML
Plain-language description of what changed and why
Which routes, which CSS classes, structural or purely visual, one screenshot or many. It always surfaces the comparison URL.
Read · Step 5: summarize
meticulous agent submit-feedback
Sends one short note to the Meticulous team with an outcome of helped, neutral or hindered, tagged with the skill name. The MCP tool submit_feedback does the same.
Runs code · Step 6: send feedback

From the official docs

Permissions

Medium risk

Claude runs the Meticulous CLI to replay a session against your app and writes files under ~/.meticulous. It also sends one feedback note to Meticulous at the end.

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

It sends a feedback note to Meticulous
Step 6 tells Claude to submit one note to the Meticulous team after the task. Say so if you do not want that. github.com
It replays a browser against your URL
The simulate command replays the recorded session in Chromium against the app URL you give it. Use your own dev server. github.com
Replay files land in ~/.meticulous
Replays, screenshots, diffs and downloaded test runs are stored there and can contain pages from your app. github.com
The plugin route also connects a hosted MCP server
The repo's .mcp.json adds the Meticulous server at app.meticulous.ai/api/mcp. The README says the plugin connects it automatically and you authenticate in your browser. github.com
The plugin route installs all 11 skills
Counted from the repo's file tree. The plugin manifest lists no hooks, and the repo has no agent definition files. Use the skills CLI route if you want to pick skills. github.com

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details

Turning on read-only	SKILL.md defines no read-only mode.
Talks to	app.meticulous.ai. The app URL you pass to --appUrl
Sign-in and keys	Uses the Meticulous CLI sign-in (OAuth token on disk, or an API token).
Scopes	Claude runs meticulous simulate, download and agent submit-feedback in your shell Replays, screenshots and diffs are written under ~/.meticulous

Questions

What does the Meticulous Simulate and Diff skill do?

It makes Claude replay one recorded session against your app and report visual and HTML differences against an earlier replay, or save screenshots for you to check.

How do I install it?

Use the install box at the top of the page. It installs the whole Meticulous set, because this skill calls meticulous-cli-update and points to the CLI skill's reference. You also need the CLI. We store no Claude app upload because it drives a local CLI.

Is this the official Meticulous skill?

It is in Meticulous's own repo, and the plugin manifest names Meticulous as the author.

What do I need before running it?

A session ID. An app URL is optional: leave it out to replay against the original recorded URL. A base replay ID is optional too, and without one nothing is compared.

Does it send my data to Meticulous?

Replays use your Meticulous account. The last step sends a short feedback note to Meticulous through the CLI or MCP tool. SKILL.md does not say what else is uploaded, so check the Meticulous docs.

Is it safe?

It bundles no scripts. Claude runs the meticulous CLI, which opens your app URL in a headless browser and writes replay files under ~/.meticulous. Point it at your own dev server and approve the commands.

meticulous-simulate-and-diff vs meticulous-review?

This skill checks one session on demand. meticulous-review works through a whole CI test run and records decisions on its diffs.

Information

updated Nov 2, 2026

Repository: alwaysmeticulous/skills
Folder: skills/meticulous-simulate-and-diff

alwaysmeticulous/skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 7

README · github.com
Meticulous docs: setting up for agents · app.meticulous.ai
Cursor docs: Plugins · cursor.com
meticulous-simulate-and-diff SKILL.md · github.com
simulate.md · github.com
.mcp.json · github.com
marketplace.json · github.com