Skip to main content
    SolutionsRemediation

    Remediation that stays open. The fix is not the part that is missing.

    Security findings, dependency and version upgrades, quality and compliance items. Identified, triaged, owned, and still waiting on capacity. For most of them the answer is already written down. What runs out is the capacity to make the change, review it and evidence it.

    Reign Factory is in beta and being productized. It is not generally available; beta access is agreed with us and scoped in writing before anything runs.

    What arrives at the gate gate unchanged
    What you get today What the Factory sends A backlog report A list of findings stops here nobody can review a list One finding, one change What was tried, and by what Your reviewer unchanged. Same policy, same branch rules, same named approver merged Reign does not move your gate. It changes what reaches it.
    The reviewer, the policy and the approver are the ones you have. What changes is that the work arrives in a shape they can act on.
    The cost

    An open finding is not a queue entry. It is a position somebody signed.

    Remediation behaves differently from feature work. A feature that slips is late. A finding that waits is an exposure the organization has agreed to carry, and somebody has to sign that acceptance again at the next review.

    01The exposure is accepted while it waits
    The backlog is a register of decisions, not a list of tasks.

    An open finding is a position the organization has taken. Someone signed it, and at the next review someone signs it again. That is a different object from a ticket, and it is the reason the backlog does not behave like the rest of the plan.

    AskWhose name is against the findings you carried into this quarter?

    02The constraint is capacity, not knowledge
    Triage has already produced an answer for most items.

    Bump the version, replace the call, add the check, correct the configuration. What is missing is the hands to make the change, the reviewer time to look at it and the record that shows it happened.

    AskHow much of your backlog is waiting on a decision, and how much is waiting on hands?

    03Evidence is part of the work, not after it
    The change is minutes. Assembling who reviewed it, on what basis and against which finding is the afternoon.

    We can find no study that puts a number on this, and we are not going to invent one. What we can say is structural: the regulations in the next card ask for a dated trail per finding, and a trail assembled afterwards costs a senior engineer more than the change did.

    AskIf someone asked how a given fix was reviewed and by whom, how long would the answer take?

    04The clock got shorter this year
    CISA replaced the KEV directive in June 2026, and the top tier is now three days.

    BOD 26-04 superseded BOD 22-01 on 10 June 2026 and prices remediation by risk rather than by catalog: three days plus mandatory forensic triage at the top, fourteen days for most KEV items, sixty for the rest. FedRAMP applies the same rules to cloud services from 7 December 2026. Whatever your current cycle is, it was set against the old one.

    AskWhat is your current committed window for a KEV item, and when was it agreed?

    What is actually measured

    Four things are known about backlogs like yours. None of them are ours.

    Every figure below is somebody else’s, named with its date and its population, because a page that argues about your backlog should not be the one inventing the numbers. iTmethods publishes no figure for an estate it has not examined.

    What is measuredWhose figure, and whenWhy it is on this page
    26% of CISA KEV vulnerabilities were fully remediated in 2025, down from 38%. Median time to full remediation rose from 32 days to 43.Verizon, Data Breach Investigations Report 2026, May 2026. More than 22,000 confirmed breaches across 145 countries.Every KEV item has a known fix by definition. Three quarters of them still did not close. That is the capacity argument with the knowledge argument removed from it.
    Organisations fix a median 10% of their flaw backlog per month. 82% now carry flaws more than a year old, up from 74%.Veracode with the Cyentia Institute, State of Software Security 2026, 24 February 2026. 1.6 million applications, 141.3 million findings.The drain rate is below the fill rate, and it is getting worse rather than better. A backlog on those numbers does not clear by working harder at it.
    For 95% of downloads of a known-vulnerable component, a fixed version already existed.Sonatype, State of the Software Supply Chain 2026, 28 January 2026. Registry telemetry across 9.8 trillion downloads.This is the closest thing to a measurement of “the fix is usually known.” It is not a claim that the fix is easy — see the upgrades page for the other half of that.
    CISA BOD 26-04, issued 10 June 2026, supersedes BOD 22-01 and BOD 19-02.CISA, 10 June 2026. US federal civilian agencies, and every cloud service selling into them via FedRAMP from 7 December 2026.It is the reason a remediation window agreed in 2025 is probably the wrong window. Any supplier still quoting the fifteen-day KEV rule is quoting a revoked directive.

    We read the BOD 26-04 remediation tiers through Tenable and Cybersecurity Dive, which agree with each other; CISA’s own pages would not open to us directly, and we would rather say so than imply we read the primary text. Everything else above was taken from the publisher. None of it is an iTmethods measurement and none of it is a prediction about your estate.

    How the work runs

    One delivery path, pointed at your findings.

    AI coding at scale is the foundation capability of Reign Factory. Remediation, upgrades and waiting delivery are the same path pointed at different populations of work. The change arrives as a merge request in your own repository, in front of your reviewer, and it stops at your gate.

    01 Agree the population Named findings, in named repositories, with the rules that make an item eligible to be attempted. In writing
    02 Eligibility decides scope The rules decide what the Factory is permitted to touch. An item outside them is not started, and the rules widen only when both sides want them to. Yours to set
    03 A named engineer Embedded for this population, running the work with your team and staying with it. Not a queue
    04 A merge request in your repository Proposed under your branch rules, against your review policy, in the repository you already run. Your repo
    05 Checks report, they do not gate Scanners report against the change. No check confers permission to merge. The finding list is an input to the work, not a gate on it. No auto-merge
    06 What cannot be completed is handed back With what was learned attached. It is not forced through, and it is not left half-made in a branch nobody owns. Refusal recorded
    07 Your reviewer decides Reign Factory does not merge, does not release and does not decide what ships. Named approver

    The only figures we publish about this way of working are our own. Across eight of our own repositories, 326 merge requests were merged. Median time from merge request open to merge was 19 hours. 326 of 326 merged with a recorded approval from a named iTmethods engineer. Measured on our own engineering, 11 July to 11 August 2026, GitLab, release commits excluded, internal environment only, two-person team. The method is written up in full at what we measured. It is not a customer result, and we do not present it as one.

    Where the record fits

    The record of how a change was made is written while it is being made.

    Every model call the Factory makes on the way to a merge request passes Reign Gateway. That is where the identity, the policy, the model and the decision are recorded — at the point the call happens rather than reconstructed from it afterwards. It is the same boundary the assistants your engineers already use pass through, which is why one policy covers both.

    What the record carries
    For each call: the identity that made it, the policy that applied, the decision taken, the model that answered, and the time. Joined to the merge request it produced, so the question “how was this fix made, and who approved it” has an answer that already exists when it is asked. A refusal is recorded as a refusal; a failure is recorded as a failure, with no fallback behind it.
    The part worth being precise about. This is a record of how a change was produced, prepared so a reviewer can follow it. It is not a judgment that the change is correct, and it does not replace your review. iTmethods makes no compliance, certification or accreditation claim under any framework. How the boundary works → · Regulatory alignment →
    Shared responsibility

    Who holds what, before the first finding is attempted.

    Written down before anything runs, so the boundary is a document rather than a discovery halfway through the first population.

    WhoWhat they hold
    Reign Factory, automatedAttempt the eligible items. Open one merge request per finding or per closely related set. Run the checks the repository already runs and report what they said. Carry the record of what was tried, by what, and under which policy.
    Customer authoredWhich findings are in the population and which repositories are in scope. The eligibility rules. Your branch, review and release policy, unchanged. Every merge decision. What an acceptable risk position is, and who signs it.
    Operating partner engagementThe named engineer embedded for the population. Triage of what came back declined and why. The periodic review of whether the rules should widen, narrow or stop.
    Before the scoping call

    The questions that arrive every time.

    Can we buy this today?
    Reign Factory is in beta and being productized. It is not generally available; beta access is agreed with us and scoped in writing before anything runs. What that means in practice is that the population, the eligibility rules and the engagement are agreed before anything is attempted, and either side can stop.
    You already patch our platforms. How is this different?
    Different layer, different population. If iTmethods operates your GitLab, Bitbucket, Jira or Camunda under Reign Ops, patching that substrate on an upstream-matched cadence is part of that service and there is nothing extra to buy. This page is about findings in your own applications, in your own repositories, which nobody patches on your behalf because nobody else owns the code. Reign Ops holds the substrate; Reign Factory works the code that runs on it.
    Does the Factory merge anything?
    No. It opens merge requests. Your reviewer decides, under the branch rules and review policy you already run, and every merge carries a named human approver. Autonomy is off by default and is turned on per population, by you, in writing.
    What happens to a finding it cannot fix?
    It is handed back with what was learned attached, and it is recorded as declined rather than attempted. Some findings need a design decision, a product owner, or a conversation with the team that owns the interface. A clear hand back is worth more than a change nobody wants to review.
    Do the scanners decide what gets merged?
    No, and we would treat it as a problem if they did anywhere in your pipeline. Scanners report. The finding list tells the Factory where to look and tells your reviewer what the change was for. It does not confer permission and it does not stand in for a human approval.
    Will you tell us how much of our backlog will close?
    No. That is your measure, taken in your environment, and we would rather agree it with you before we start than assert it now. The only figures we publish about this way of working are our own, measured on our own repositories, and they are stated in full at what we measured.
    Which deployment shapes can we have?
    A single-tenant dedicated instance, in your own AWS or Azure account or in infrastructure iTmethods operates. Both are available today; Google Cloud is planned for 2027. Air-gapped and sovereign are in development. There is no multi-tenant or shared option at any tier. Deployment options states which is which without softening any of them.
    AWS Advanced Tier Services Partner SOC 2 Type II on Reign Ops
    AWS Advanced Tier Services Partner and Validated Managed Service Provider. Twenty-one years operating regulated engineering environments.
    Next step

    Bring us the backlog you have stopped forecasting.

    Show us a population of findings you consider stuck, and we will tell you which parts we think this path suits and which parts it does not. That answer is free and it is sometimes no.