UAE · ENBook a demo
Procure AIEngineeringAssistant

Security Scanning Agent

Scans code and systems for vulnerabilities.

Security Scanning AgentPlatform and Reliability
Industry
Procure AI
Department
Engineering
Team
Platform and Reliability
Kind
Assistant
Approver
Security Lead
Code
ENG-16

It runs and reads static analysis, dependency, secret, container and configuration scans across the platform, the supplier portal and the integration adapters, removes false positives and duplicates, and rates each finding by exploitability and exposure, with extra weight on anything that reaches supplier bank details, payment files or cross-tenant data. It opens a private fix ticket per confirmed finding and sends critical findings, exposed secrets and risk acceptances to the security lead.

How it works

6 steps, and a person signs off before step 6.

  1. 01Looks it up

    Finds the record the request is about in the vulnerability scanner: the latest scan results, their history and the asset inventory entries for the targets named..

  2. 02Reads the request

    Works out what is being asked and what a complete answer needs.

  3. 03Checks the facts

    Checks figures, dates and rules against the source records, not from memory.

  4. 04Works it through

    Prepares the draft, match, recommendation or reply, with its sources.

  5. 05Asks a person

    Hands the work to Security Lead with its reasoning before anything goes out.

  6. 06Acts

    Once approved, it acts in the issue tracker: it creates a private fix ticket per confirmed finding in the owning team's queue. and records what was done.

What it works with

Your knowledge, your systems, your people.

Knows

The security policy aligned to ISO 27001 and SOC 2 controls: the severity scale and fix deadlines, the secure coding standard, the asset inventory with exposure and data classification (bank, tax, pricing, personal), client contractual security clauses and the accepted-risk register.

Reads from

the vulnerability scanner: the latest scan results, their history and the asset inventory entries for the targets named.

Acts in

the issue tracker: it creates a private fix ticket per confirmed finding in the owning team's queue.

Hands back

A list with one entry per finding, each with id, title, location, scanner, severity (critical, high, medium or low), exploitable (true, false or unknown), exposure, data_at_risk (bank, tax, pricing, personal or none), false_positive (true or false, with why), owner, fix and due (YYYY-MM-DD or null).

Asks a person when

A finding is critical or an exposed secret, it proposes accepting a risk, an active scan would run against production, or a finding may need a client breach notification.

ApproverSecurity Lead

Settings

Your limits, set by your managers.

SettingDefaultWhat it controls
Fix deadlines by severity (days)critical: 7, high: 30, medium: 90, low: 180Days from confirmation to fix for each severity.
Lowest score that gets a ticket4.0Confirmed findings below this are reported but not ticketed.
Where active scans may runstaging_onlyActive scanning sends attack traffic; production needs the approver even when allowed.
Revoke exposed secrets itselfOffWhen off, an exposed secret, such as a client ERP or bank-file credential, is reported to the security lead to rotate.
Longest risk acceptance (days)90An accepted risk is reviewed again after at most this long.

Limits are settings, not code: a manager changes them without a developer.

Rules it always follows

  • Confirm each finding before rating it: drop false positives with the reason, and merge duplicates across scanners.
  • Rate severity from the policy scale, raised one level when the asset is internet- or supplier-facing or holds bank, tax or pricing data, lowered when the code path is unreachable.
  • Any flaw that could let one client tenant read another's data, or alter a supplier's bank details or a payment file, is critical.
  • Ticket confirmed findings scored 4.0 or more, due by critical: 7, high: 30, medium: 90, low: 180, in the owning team's private queue.
  • An exposed secret is treated as compromised: revoked when off is true, otherwise sent to the Security Lead at once.
  • Run active scans only where staging_only allows, and never during a client's payment run.
  • Accept a risk only with the Security Lead's agreement, for at most 90 days, checked against client contract security clauses.

See the Security Scanning Agent agent on your own work.

We’ll run it on a sample of your work and show you what your approver would see.

Security Scanning Agent · Procure AI | Basinwright Industries