ArgueSecure 2.0.0

A complete rewrite: collaborative security risk assessment with real numbers, live editing and reports you can hand to a client.

ArgueSecure 2.0 shares a name with 1.x and nothing else. It is a new application on current foundations (Laravel 12, PHP 8.3, Vue 3), and no 1.x data is carried over.

What you can do

  • Put numbers on risk. Give each risk an impact (the loss each time it happens) and each attack a likelihood (how often per year, which may be more than once). ArgueSecure works out the expected annual loss three ways: with no defences, with what is implemented today, and with everything planned. Shelved defences never count.
  • Honest results. Results are amounts in your currency with a reduction percentage, never low, medium or high. A value you have not filled in is flagged and left out of the total, never silently counted as zero.
  • Model a web, not a tree. An attack can belong to several risks and a defence can counter several attacks. How effective a defence is can differ from one attack to the next, so strength is set on each link. Hovering a shared item highlights everywhere it appears.
  • Edit in place. Add items inline and start typing; a tabbed window holds details, values, links, references, comments and history without leaving the assessment. Drag items to reorder or move them.
  • Work together live. See who else is in an assessment, watch changes arrive, and see which item someone is editing so you never overwrite each other.
  • Reports. A print-ready report ranks defences by what they save and by the opportunity they offer, and exports to CSV and JSON. PDFs come from your browser’s print dialog.
  • Snapshots. Freeze an assessment together with its computed results, so recalibrating later cannot change a figure you already reported. Compare any two snapshots.
  • Sharing. Share with named people in a role, publish a read-only link, or make an assessment public, which grants read access only.
  • Templates and a catalogue to start from, comments, per-item revision history with revert, external references (such as CVE identifiers) and a single undo history.
  • Accounts. Administrators create accounts one at a time or in bulk from a CSV file. There is no public registration.

Security

  • Every assessment is checked against the same access rules, wherever it is reached from. Real-time channels are authorized on the server.
  • There is no default account and no built-in password. The first administrator is created on a one-time setup screen.
  • Nothing loads from outside your server, and a strict content security policy is enforced.

First start

The application key and the real-time keys are generated on first start and saved under your data directory. Back that file up: losing it signs everyone out and invalidates encrypted cookies. Set APP_URL, DB_PASSWORD, DB_ROOT_PASSWORD and DATA_PATH in .env. On first open there is a setup screen where you create the first administrator.

Upgrading from 1.x

There is no data migration: 1.x had no quantification, so an imported assessment would arrive with every value missing and every total zero.

  1. Run 2.0 alongside 1.x on a different port and data directory (see the README of the current release for installing).
  2. Create the accounts and re-enter the assessments still in use, adding impact and likelihood figures as you go.
  3. Switch your reverse proxy over. Keep the 1.x data until nobody needs it.
  4. Do not reuse any 1.x password or application key. Treat them as compromised.

Known limits

  • Moving an item under a parent that has no children yet has no drag gesture.
  • Shift+Tab does nothing in the outline: an item’s level is its type, which never changes.

View these notes on GitHub

Run this release

The self-hosting page has the commands to install or upgrade, and the demo shows the current release.