Preflight records real executions from your client's n8n instance, replays your edit against them, and compares every node's output. n8n reports that the run succeeded. Preflight reports whether the behavior changed.
An actual catch from live testing: one changed word, zero errors, wrong output nightly.
They run “successfully” while doing the wrong thing, and surface days later — on the client's side. Each of these is a recurring topic on the n8n community forum.
One node changes, the editor looks fine, and a downstream output quietly changes shape. Data is wrong for a week before anyone notices.
A third-party API renames a field, a token expires, a node is deprecated in an update. Most incidents aren't caused by the person who answers for them.
A sub-workflow is edited; the parent's definition is byte-for-byte identical. Definition-based diffs report no change at all. The behavior changed.
Connect a client instance with its base URL and an API key — nothing is installed on their side. From then on, every change follows the same four steps.
Real executions become replayable fixtures: per-node inputs and outputs, with an honest completeness score. Personal data is redacted on ingest.
Your edited workflow runs in Preflight's sandbox against the recording, external calls served from the fixture. Client credentials never leave the client instance.
Every node's output is compared, with noise rules for timestamps and volatile fields. A red diff names the exact node — and the cause: your edit, or an outside API that no longer answers the way it did when recorded.
Deploys require a green or explicitly approved replay of that exact change. The previous version stays pinned for one-click rollback, and every deploy is audited.
Every n8n release produces the same forum thread: workflows that ran flawlessly for months, crashed after the version bump. Preflight replays each client's recorded executions on the target n8n version in its own sandbox — the client's server is never touched — and reports, workflow by workflow, what survives. Each workflow also replays on the current version, so one that was already unstable is called flaky instead of being blamed on the upgrade.
n8n's Business tier compares workflow definitions. Preflight replays and compares workflow outputs — the only way to catch a sub-workflow edit, an expression silently resolving to nothing, or an API that changed its response format.
| n8n Community | n8n Business (~€667/mo) | Preflight | |
|---|---|---|---|
| Version history & rollback | — | Included | Included, zero instance setup |
| Definition diffs | — | Included | Included |
| Replay real executions against an edit | — | — | Included |
| Per-node behavior diffs with cause analysis | — | — | Included |
| Pre-upgrade check on the target n8n version | — | — | Included |
| Deploy gate & audit trail | — | — | Included |
| Agency pricing model | — | Per instance | Per agency, from $29/mo |
Per-instance pricing is precisely what agencies resent. One subscription covers your client base.
Founding members: join the waitlist now and keep these rates for the life of your subscription.
Preflight talks exclusively to n8n's public API. Nothing is installed on client instances and no n8n internals are touched.
Client credentials never leave the client instance. Replays run entirely from recorded stubs.
Nodes without a complete recording are marked untestable. A test never claims coverage it doesn't have.
Every deploy writes its audit entry before it executes — who, what, when, and which test cleared it.
Built for self-hosted n8n, Community edition included — where agencies actually run client work.
Personal data is redacted when recordings are ingested; raw values are sealed in an encrypted vault.
This isn't a mockup: the engine — record, replay, diff, gate, deploy, rollback — is built and running against live n8n instances today. The beta opens to the founding cohort first, at founding rates — kept for life.