In a measured replay of 193 real Cloudflare Workers SDK commits from repo history, the same Wrangler deploy checks passed in standard CI every time. DriftFence still flagged the same deploy change across the full replay window.
With approved deploy behavior fixed in Git, DriftFence would have kept surfacing the same change while the repo's own checks still passed.
The first commit in this window already failed the DriftFence check, and DriftFence kept flagging the same Wrangler deploy scenario across the rest of the replay window. DriftFence flagged the change at the operation level even though the same checks still passed in standard CI.
Across that replay window, the same Wrangler deploy behavior
kept changing: the baseline expected error code
10007, but later commits returned no error code.
DriftFence caught that repeated change every time.
The remote-secret conflict warning and remote-config diff scenarios did not trigger DriftFence on the same commits. The flagged change stayed on one deploy surface.
This window does not have a before-and-after turn like Firebase. The first commit already triggered the DriftFence check. What matters is that the same deploy change kept showing up while the same deploy checks stayed green and two nearby deploy scenarios did not trigger DriftFence.
This replay fixes one Wrangler deploy baseline, reruns the same deploy checks on each commit, and counts the recurring signal once even though it persisted across the full window. Exact commit references are linked below.
193 commitsThe links below show the benchmark notes and pinned replay plan for this Cloudflare example.
The benchmark results log records the completed Wrangler deploy window, including the replay size, outcome mix, and representative first flagged change.
The Workers replay definition fixes the three Wrangler deploy workflows and the replay window used here.
See the product, results, pricing, and trust pages alongside this Cloudflare example.