API security

API penetration testing for REST, GraphQL & microservices

Test REST and GraphQL surfaces, auth boundaries, and rate limits with evidence suitable for DevSecOps teams.

Infronest product screenshot (apivapt)
In short

What are BOLA and BFLA, and do you check for them?

BOLA (Broken Object Level Authorization) is when a user can read or change another user’s object by changing an ID; BFLA (Broken Function Level Authorization) is reaching an admin-only function as a normal user. The authz-matrix scanner replays requests across roles specifically to surface both.

Product proof

Already live in the product

Infronest module
Protected app route: /vapt-integrations

This engagement can align with shipped product capabilities in your tenant workspace.

Platform

Included in the Infronest workspace

These capabilities ship in the product today—consulting adds human validation and attestation where required.

  • CI & integrations
Infronest product screenshot (apivapt)
Capabilities

What you get

Delivered with AI-assisted engineering and human review on architecture, security, and production readiness.

  • API security scan type in platform
  • Schema-aware fuzzing concepts
  • CI integrations and webhooks
  • Broken object level authorization checks
  • Export for SIEM or ticketing
Infronest product screenshot (apivapt)
Workflow

What teams can do here

  1. Step 1Import API spec
  2. Step 2Map auth roles
  3. Step 3Execute test suite
  4. Step 4Track fixes in Git

How it works

  1. 01

    Discover the surface

    Import an OpenAPI/Swagger spec or let the discovery scanner enumerate common REST and GraphQL paths. Knowing the real endpoint list keeps testing scoped and thorough.

  2. 02

    Map the auth roles

    Provide sessions for the roles you want compared — for example an admin and a standard user — so authorisation boundaries can be tested rather than assumed.

  3. 03

    Run the API and authz checks

    The API scanner looks for unauthenticated access to sensitive data (CWE-306), IDOR/BOLA (CWE-639), missing rate limiting (CWE-770) and unsafe methods (CWE-650); the authz-matrix scanner replays requests across roles to catch BOLA and broken function-level authorisation (BFLA). GraphQL and JWT/OAuth scanners cover those surfaces.

  4. 04

    Verify and report

    High-impact findings are confirmed by hand, scored with CVSS 3.1, and exported in one report. Results can also feed CI via the integrations hub.

“

A worked example

A typical scope: a B2B SaaS with 85 REST endpoints and a GraphQL gateway hands over its OpenAPI spec plus two tokens — tenant admin and basic member. The authz-matrix replay finds GET /api/v2/invoices/{id} serving another tenant’s invoice to the member token, the API scanner flags an unauthenticated /api/export returning PII, and the login endpoint shows no rate limiting. An analyst reproduces the cross-tenant read, attaches the captured request as evidence and scores it high. Once an ownership check and throttling ship, the suite re-runs and the report shows all three findings closed.
Both. There is a dedicated REST API scanner and a separate GraphQL scanner, alongside JWT and OAuth checks for token-based auth. The right combination is chosen from your endpoint inventory.

Find the BOLA before a customer does

Role-matrix replay across your real endpoints surfaces the cross-tenant reads and unauthenticated exposures that standard scans miss — with evidence attached to every confirmed finding.