You want to point Claude at a Notion database and say normalize the tags on every row, tighten each summary to one line, and standardize the rich-text after this long edit cycle. What stops you is everything that reads from it. Views, linked databases, and any Notion site the table drives all update the moment a row changes, and the API gives you no diff and no undo.
Scratch changes where the edit happens. Claude does 99% of the work, reading and rewriting every row as files on your laptop. The last 1%, deciding what actually writes back, stays with you. Nothing reaches the live database until you have seen the change as a diff and approved it.
How it works
- Scratch pulls your database into files. A Notion database comes down to a folder on your laptop, one file per row, every property laid out for editing.
- Claude edits the rows. Open the folder in the Claude desktop app. Try a prompt on a few rows in Chat, then let Cowork or Code run it across the whole table. Normalize the tags on every row and tighten each summary to one line. Claude edits the files, never the live database.
- You review every diff and publish. Scratch shows each changed property next to the original, word by word. Approve what ships, and Scratch writes only those rows back through the Notion API.
What people use it for
Most people arrive with a database that has gone messy, because tidying it by hand means opening every row one at a time.
- Normalize tags and categories across every row.
- Tighten each summary or description down to one line.
- Standardize a rich-text column after a long, messy edit cycle.
- Clean up the database that drives a Notion site so the site reads cleanly.
- Fill empty select or status cells by reading each row.
Pull a handful of rows to feel the loop, then point Claude at the whole database.
Why not an MCP server?
A Notion MCP server or integration hands Claude a direct line to your live workspace. The write button is wired straight in, so one bad pass rewrites every row at once and whatever those databases drive updates instantly.
Scratch gives Claude the same full read and write access, but against a local copy. The write-back step is pulled out and handed to you. Claude can change anything; only you can commit it. On a database your team works in every day, that is the difference that matters.
What Claude edits in Notion
- Database properties: titles, rich-text, select, multi-select, and dates
- Relation, people, and files properties
- Cover and icon metadata
Page bodies stay read-only. The connector edits the database side, not the block trees, toggles, or embeds inside each page, and computed properties (formula, rollup, created and edited metadata) stay locked. For the full picture, see Scratch for Notion.
Questions people ask
Is this an MCP server or a Notion integration?
No. An MCP or integration gives Claude the write button straight to your workspace. Scratch does not. Claude gets the same access, but writing back is a separate step you approve, one row at a time.
Will Claude touch my page content?
No. Claude edits database properties only. Page bodies, block trees, toggles, and embeds are never exposed for editing, and computed properties like formulas and rollups stay locked.
Can I undo a change after it writes back?
Yes. Every written row is reversible from Scratch, per row. The original sits next to the rewrite until you decide which one stays.
Why not just bulk-edit with a CSV or a script?
A CSV import and a custom script both write straight to the live database with no diff and no per-row approval, and a find-and-replace only does exactly what you spelled out. Scratch puts an agent on the edit, so it handles the messy cases a rule cannot, and holds every change as a word-level diff you approve before anything ships.
Can it handle a big database at once?
Yes, that is the use case. Pull a handful of rows to feel the flow, then pull the whole database. Cowork and Code are built for database-scale jobs.
Do I need to be technical?
No. Install Scratch, connect Notion, point Claude at the folder, and approve the diffs.
See it on your own database
The fastest way to trust it is to watch it run on your data. See it run on your Notion database →, or download Scratch free and run the first pass yourself.