CloneBrake — when a store gets copied
Project history
The decisions, releases and learning that shape this project. Expand a milestone to see the rationale and evidence.
8 milestones · Newest first
2026
· Product decisionTightened the report release checksRequired human-confirmed copies and visible matched evidence before an outbound report can be released, with an explicit review of warnings.Expand decision details
Why this changed
Shared supplier photos and legitimate stockists can resemble copies. Keep that uncertainty visible before making a claim to a store owner.
Evidence & context
The revised handoff adds a block for missing matched photos or copied words, warnings for established domains, and a preview using the actual customer report component. These are design requirements; their implementation and effect on false positives have not been independently verified here.
· Product decisionSeparated outreach from service communicationsChose a dedicated outbound mailbox, separate from receipts and refund messages, with send-time checks and reply handling.Expand decision details
Why this changed
Keep outreach failures from disrupting case communications and make the mailbox integration portable through standard email protocols.
Evidence & context
The requirements specify SMTP and IMAP, suppression checks, pausing follow-ups on replies and stopping on authentication or policy blocks. This records the operational design, not verified sending reliability or conversion results.
· Product decisionScoped a small first outreach trialDefined a first batch of ten individually checked outreach emails, followed by manual handling of the first paid cases.Expand decision details
Why this changed
Test demand and operating effort before committing to a larger launch, with separate readiness checks before outreach and before accepting payment.
Evidence & context
The launch checklist sets out the trial and its prerequisites. It does not establish that the emails were sent, customers paid or cases were resolved.
· Design decisionPut the evidence before the agent’s recommendationRevised the report and operator review around side-by-side evidence, plain-language explanations and a separate decision panel.Expand decision details
Why this changed
Make it easier to inspect the underlying comparison before accepting the agent’s call, reducing the risk of rubber-stamping its recommendation.
Evidence & context
The approved redesign replaces customer-facing scores with explanations of what was copied and puts proof before the payment decision. This records the design direction, not a measured improvement in review accuracy.
· Brand decisionSelected CloneBrake after naming checksUsed Brand Launch Agent to compare names, investigate collisions and select CloneBrake; purchased the matching .com domain.Expand decision details
Why this changed
Find a distinctive name that suggests intervention without promising removal, with room beyond the initial retail use case.
Evidence & context
The naming exercise rejected Cloneward and Impostra after domain or social-name conflicts. Checks informed the choice but did not establish trademark clearance or ownership of unclaimed-looking social handles.
· Product definitionDefined a detection-first MVPExpanded the initial manual-pilot proposal into a staged plan for detection, operator review, customer reports and owner-approved complaints.Expand decision details
Why this changed
Test whether finding and explaining suspected copies creates a useful starting point before scaling takedown workflows or monitoring.
Evidence & context
The October plan separates setup, detection and reporting, outbound validation, self-service and monitoring. Stage gates are targets to test, not achieved customer or enforcement outcomes.
· Product definitionDefined a manual pilot before buildingProposed five to ten authorised cases to test whether better evidence improves incident response.Expand decision details
Why this changed
Test effectiveness, effort and willingness to pay before automating the workflow. Define stop criteria and owner approval requirements at the outset.
Evidence & context
The draft v0.1 PRD is dated 23 September and explicitly marked pre-validation. The cases, takedown outcomes and commercial assumptions remain untested.