Self-hosted vulnerability remediation
Hakeem reads your scanner's export, ranks every finding by real-world risk, and turns the top of the list into upgrades your developers can finish in the hours you have. A finding closes only when a later scan shows it's gone.
Runs on your own servers. Never touches your code. Works without internet access.
Where remediation stalls
How it works
Everyone sees only the work they can act on.
Upload a scanner export, or have your CI pipeline send a lockfile. Hakeem groups findings by the upgrade that fixes them and ranks them. Your scanner's severity stays exactly as it was. Hakeem's priority sits beside it and sets the order.
The developer gets an email from your own mail server, opens the item in Hakeem, and either picks a fix or sends it back with a note. Developers can't upload or delete scans.
When the next scan shows the fix in place, the item moves to Remediated, marked Scan Verified. Until then it stays open.
What else it does
No fix released yet, a malicious package, or a risk you've decided to accept.
When OSV records this package and version as malicious, Hakeem opens an incident and suggests a clean version to move to.
Findings with no fix yet stay on the list and join the plan as soon as a fix is released.
New advisories are checked against the packages Hakeem already knows each application uses, so an application that hasn't been scanned lately is still covered.
Wiz and Trivy exports, generic JSON and CSV, and lockfiles from CI: package-lock.json, poetry.lock, requirements.txt, pom.xml, go.mod, Gemfile.lock.
Admin, analyst and developer. Teams own applications, and developers only see their own team's work.
KEV from CISA, CVSS from NVD, EPSS from FIRST. Where a scan carries its own copy of a score, Hakeem uses the original source instead.
The Record
Every accepted risk keeps its reason and a review date. Sign-ins, credential changes, hand-offs, risk acceptances and refused requests go into an audit log where each entry is linked to the one before, so any tampering shows. The screen tells you whether the chain is intact.
Self-hosted and air-gapped
Hakeem is installed on your own infrastructure and keeps everything in its own database. On a network with no internet, turn on offline mode: it makes no outside requests, works from the catalogue copies shipped with each release, and shows you how old they are.
For procurement and the CISO
Who it's for
With forty repositories and no auditor, a pull-request bot is probably enough. Hakeem is for teams that answer for what they fixed, what they didn't, and why.
Pricing
A resource is each application or asset your scanner reports.
| Resources | 50 | 200 | 300 | 500 | 1,000 |
|---|---|---|---|---|---|
| Per year | $6,000 | $22,500 | $32,500 | $52,500 | $97,500 |
Questions
Scoring is only the first step. Hakeem then gives each fix an owner, fits the work to your team's hours, checks the next scan to prove it's done, and records every exception.
No. It reads your scanner's export. The scanner's severity is shown as the scanner wrote it, and Hakeem's priority sits beside it.
When a bot merges an upgrade, the next scan sees it and the item is remediated. The cases that stay manual are the ones Hakeem is for: no fix exists, the fix breaks something, the package is malicious, or the risk is accepted on purpose.
No. Hakeem takes no source-control credential and never writes to a repository.
As current as the catalogue copies it holds, and each one's date is on screen. A signed advisory store can be installed between releases. KEV and EPSS update with each release.
We'll walk you through it with a scanner export.
Usually within one working day.