Diagnostic only · 2026-08-06
Dark Factory: quarantine integrity incident
- Stopped at
- review
- Reason category
- policy
Executive summary
The scheduled run stopped at the research integrity boundary when the signed quarantine receipt did not bind to the supplied sanitised longlist. No candidate was created or exposed, and this diagnostic preserves the required remediation without revealing unsafe research material.
The factory story
The brief
The scheduled run began with the isolated research boundary required before any game direction, candidate, or release work could proceed.
What was researched
Only sanitised evidence records were considered. The supplied packet was rejected because its signed receipt did not bind to that exact packet.
What was decided
The mismatch was treated as an integrity incident, preserved unchanged, and withheld from every candidate-dependent stage.
What was built
No game, asset, design, build, test candidate, or public game route was created by this run.
How it was validated
Containment recorded the rejection and verified that ordinary public routes remained available without exposing a run candidate.
Independent review
The controlled disposition classifies this integrity failure outside automatic repair and restart authority for the supplied packet.
What happened at release
No candidate was merged, deployed, catalogued, linked, or made playable. This is a diagnostic-only report, not a candidate release.
Outcome
Production content remained unchanged by the incident. A fresh isolated capture is required before a future candidate pipeline can begin.
What we learned
A signed quarantine receipt must bind to the exact sanitised packet before trusted processing begins, and unavailable token counts must remain explicitly unavailable rather than estimated.
Run timing and token use
Step timing and token telemetry was introduced after this historical run, so no figures are available to publish.
Checks completed
- Research sanitisation rejected the mismatched packet before candidate creation.
- Containment preserved the incident evidence without candidate, catalogue, or deployment mutation.
- Ordinary public routes were verified available without a run candidate exposure.
- The terminal incident diagnostic identifies the fresh capture and dependent reruns required.
What happens next
The next calendar day’s run remains scheduled.