Support

Scorly is free and open source, maintained in the open on GitHub. The fastest way to get help, report a bug or ask for a feature is to open an issue — it's public, it gets a reply, and the fix ships in a version anyone can read.

Reporting a bug

Scorly runs against whatever page you point it at, so almost every bug comes down to a specific page behaving in a way the analyzer didn't expect. Include as much of this as you can:

• The URL of the page you were analyzing, if it's public. If it's localhost or a client site, a description of the relevant markup is enough — or the exported JSON with anything sensitive removed.
• Which tab you were on, and what you expected it to say instead.
• Your browser and version, and the Scorly version from the About tab.
• A screenshot of the panel, if something looks wrong rather than reads wrong.

The Export buttons in the popup footer produce a JSON file with the full analysis in it. Attaching that to an issue usually tells the whole story.

Common questions

Scorly's score differs from another SEO tool's. Which is right?

Both, probably. There is no official on-page SEO score — every tool invents its own weighting, so two of them disagreeing is expected rather than a bug. What matters is the individual checks underneath the number, which name exactly what passed and failed. Every rule Scorly applies is readable in popup/scoring.js, so you can see precisely why a page scored what it did.

Do the "AI" features send my page to an AI service?

No. Nothing is sent anywhere. The AI Insights tab, the AI Visibility score and the AI-copy and vibe-code detectors are all local pattern matching — counting phrases, class names, attributes and punctuation in the page already open in your browser. "AI" refers to the audience (how readable your page is to AI answer engines), not to the method. See the privacy policy for the detail.

Why does it need access to all sites?

Because it can analyze whatever page you're on, and that could be any URL — including localhost and private-network addresses. The permission is only ever used on the tab you're actively inspecting, when you click the icon. Scorly doesn't run on a schedule and doesn't read pages in the background.

It says my page is client-rendered, but it looks fine to me.

That check compares the HTML as the server sent it against the DOM after JavaScript has run. If the content only appears in the second one, a crawler that doesn't execute JavaScript sees an empty shell — which is the point of flagging it, even though the page looks perfect in your browser.

The AI-copy detector flagged writing I wrote myself.

It counts surface patterns and names the passages it matched, and it cannot prove how anything was written — good human writing trips some of these tells routinely. That's why a page has to trip three independent families of tell before anything is deducted, and why the whole detector is capped at a small share of one category. Read the evidence it lists and judge it yourself.

Where are my saved snapshots?

In local extension storage on your own machine, never uploaded or synced. Scorly keeps at most 8 and discards the oldest beyond that. You can delete any or all of them from the compare view, and uninstalling the extension removes them.

A snapshot I saved earlier won't compare properly.

Snapshots saved by an older version can be missing fields that a newer version expects. Those sections are skipped rather than guessed at. Re-saving the snapshot on the current version fixes it.

My AI agent says Scorly isn't installed.

Run npx -y github:dazecoop/scorly status in a terminal. It lists each agent app, whether Scorly is installed there and whether it starts. Each app needs a full restart after installing, and the test prompt on the AI agents page tells the agent to search for the tool, which agents with many tools need.

Can my AI agent use Scorly without the extension?

Yes. The MCP server and CLI audit pages in a headless Chrome on their own. The extension's agent switch is only needed for audits in your own browser, with your logins. See AI agents.

Does it work in Firefox and Edge?

Yes. The same codebase builds for Chrome, Edge and Firefox. Install it from the Chrome Web Store for Chrome, Edge and Brave, or from Firefox Add-ons for Firefox.

Can I use it on a site I don't own?

Yes. Everything it reads is already public to anyone visiting the page, and nothing is stored unless you save a snapshot.

Feature requests

Open an issue describing what you'd want to see and the situation it would help with. Pull requests are welcome too — the source is MIT licensed and the scoring rules are deliberately kept readable so they can be argued with.

Contact

For anything else, including privacy questions, open an issue on the Scorly repository or contact @dazecoop on GitHub.