RITRA · KIRA

Attack surface management and automated penetration testing

KIRA automates the repetitive part of a penetration tester’s job: it discovers what you expose, tests it in a controlled way within the scope you authorise, and delivers verified findings prioritised by real-world exploitation data. Run it weekly or monthly and measure risk in the months between manual pentests.

  • 56 specialised agents
  • OWASP Top 10
  • MITRE ATT&CK
  • EPSS · CISA KEV
  • EU company · Estonia
  • Engines and AI hosted in the EU
  • Your own dedicated instance
  • Verified assets only
  • Backed by specialists

The problem

An annual pentest leaves eleven months in the dark

A penetration test is a snapshot. The week after the report, a developer deploys a new API, a test server goes live without authentication, or a critical CVE is published for software you run. Nobody looks again until next year’s test.

KIRA closes that gap. It runs the same method on the schedule you set, so you can see what has appeared, what has been fixed and what is still open.

Two scanning modules

Web applications and infrastructure, with 56 specialised agents

42 AGENTS

Web applications and APIs

The OWASP Top 10 and beyond: injection, cross-site scripting, broken access control, authentication flaws, API weaknesses, file uploads, business logic and misconfiguration. Integrates Nuclei, Nikto and WPScan.

14 AGENTS

Infrastructure and network

Network reconnaissance, ports and services, weak encryption, default credentials, known vulnerabilities, controlled exploitation and Active Directory mapping.

ATT&CK

Attack techniques

Infrastructure findings are mapped to MITRE ATT&CK techniques, so you can see how an attacker would actually use them.

M365 · AZURE

Inventory from your cloud

Read-only import of devices and resources from Microsoft 365 and Azure, leaving out inactive or duplicate entries.

Methodology

The same method as a pentester, automated

Preparation

Access to the application, crawling and technology fingerprinting.

Discovery

Agents look for vulnerabilities in parallel.

Correlation

Findings are chained into attack paths and exploitation is planned.

Targeted exploitation

Technology-specific agents, for example for WordPress.

Exploitation intelligence

Every finding is checked against NVD, EPSS and CISA KEV at the time of analysis.

Verification

Findings are confirmed, confidence is calibrated and duplicates are removed.

Reporting

Technical and executive reports in your corporate identity.

Behind it is an AI-orchestrated multi-agent engine, hosted in the EU, that chains findings and builds proofs of concept in an isolated environment. When a result is uncertain, several models review it before it can reach your report. Every finding is rated confirmed, probable or suspicious.

From finding to risk

A 0–100 risk score for every asset

Each asset gets a score calculated the same way every time, so runs can be compared and you can see what has been closed. Exposure carries extra weight: an internet-facing asset counts three times as much as one whose exposure is unknown.

FactorWeight
Vulnerabilities by severity60%
Active CVEs15%
Exposure (internet or internal)15%
High-risk ports (RDP, SMB, Telnet…)10%
ReportsTechnical and executive, in PDF, Word and PowerPoint, with your branding
ExportsExcel, CSV and a standard format for other tools
PrioritisationNVD, EPSS and CISA KEV
FrequencyWeekly, monthly or on demand

Safety by design

An offensive tool that is safer than what it tests

  • Only what is yours. Domain ownership is proven, and re-verified, before any scan. A person approves every IP range.
  • External or internal. External tests run from our EU-hosted platform. Internal networks are tested through a lightweight probe that runs no engine and opens no inbound ports.
  • Non-destructive tests. Injection tests from the probe are non-destructive and limited to the authorised scope.
  • Your findings stay with you. Findings are stored in your own instance. Our platform keeps only basic run metadata, never findings or targets.
  • Scope agreed in writing before any work begins.

Comparison

KIRA, a manual pentest or a vulnerability scanner?

Vulnerability scannerKIRAManual penetration test
FrequencyScheduledWeekly, monthly or on demandUsually annual or per major release
Finds unknown assetsOnly the targets you listYes, fed by TARA’s discoveryWithin the agreed scope
Confirms exploitabilityRarelyYes: chained findings, proofs of concept, confidence levelYes, in depth
Complex business logicNoPartiallyYes
PrioritisationCVSS scoreEPSS, CISA KEV and a 0–100 asset scoreExpert judgement
Best forPatch hygieneContinuous visibility between auditsDepth, certification and customer requirements

KIRA complements your pentest, it does not replace it. A manual penetration test is still needed for complex business logic, advanced authenticated sessions and creative attack chains. KIRA tells our pentesters where to look first, and their findings make the next KIRA run sharper.

Compliance

Where KIRA fits ISO 27001 and NIS2

  • ISO/IEC 27001:2022, control 8.8 (Management of technical vulnerabilities) requires you to obtain information about vulnerabilities, evaluate your exposure and act on it. KIRA’s scheduled runs and run-to-run comparison provide that evidence.
  • Control 5.9 (Inventory of information and other associated assets) is supported by KIRA’s single asset inventory, with services, findings and CVEs per asset.
  • NIS2 Article 21 includes vulnerability handling and procedures to assess the effectiveness of security measures. Comparable, scheduled results help show both.

Every KIRA finding flags the affected controls in RITRA vCISO, and the assets discovered by TARA feed KIRA’s inventory.

FAQ

Frequently asked questions

What is attack surface management?

Attack surface management (ASM) is the regular discovery of everything your organisation exposes — domains, services, applications, open ports — together with risk assessment and tracking over time. KIRA combines it with automated penetration testing.

Is KIRA continuous penetration testing?

KIRA repeats its tests on the schedule you choose, weekly or monthly for example. Every run uses the same method, so results can be compared over time.

Can automated pentesting replace a manual pentest?

No. KIRA automates repetitive work and gives you visibility between audits. A manual pentest is still needed for complex business logic, advanced authenticated sessions and anything that requires human judgement.

How does KIRA reduce false positives?

Every finding goes through a verifier. Uncertain cases are reviewed by several AI models, confidence is calibrated and duplicates are removed before anything reaches the report.

Could KIRA break my systems?

Tests stay within the authorised scope and injection tests on internal networks are non-destructive. No scan starts without proven ownership and a scope agreed in writing.

What reports does KIRA produce?

A technical and an executive report for every run, in PDF, Word and PowerPoint with your branding, plus exports to Excel, CSV and a standard format for other tools.

Next step

Measure your attack surface before an attacker does.

We agree the scope in writing and help you set up your first runs.