How Runny Works

What Runny does

You paste a script. Runny stores the exact bytes and serves them over HTTPS from https://runny.sh/r/{slug}. Completions, scans, and team edits happen around that file. They do not change what the raw endpoint serves until you save a new revision.

We host your code. We never execute it on our hosts.

What the client does

The machine that runs curl … | bash (or python3, pwsh, node) downloads the bytes and executes them locally, under that machine’s user, network, and secrets. Review the share page before you pipe.

Lifecycle

• SlugRandom 12-character id. Alphabet omits I, L, O, and U.
• FetchA successful raw GET counts as a script fetch. Free accounts cap fetches per day and month.
• ExpireAfter expires_at, or when max_fetches is reached, the raw URL returns 410 with expired.
• RevokeThe next raw fetch returns 410 with gone.
• ArchiveAPI DELETE sets archived_at. Raw treats that like revoke.