Skip to content
Featured on Startup Fame

Detectify’s Agentic Tooling lets AI agents test apps—then logs the proof

Detectify AB has released Agentic Tooling, a set of tools that lets organizations send their own AI agents to test live applications for vulnerabilities. To address concerns about AI reliability, the company also launched a Security Ledger—an immutable log that records every test’s evidence, written by Detectify rather than the AI agent.

Editor, Lazyfounder

Published 4 min read
Detectify’s Agentic Tooling lets AI agents test apps—then logs the proof
Image: SiliconANGLE via source

30 SEC SUMMARY

  • Detectify’s Agentic Tooling lets companies deploy their own AI agents to test running applications for vulnerabilities, then logs every test in a tamper-proof Security Ledger.
  • Eighty percent of findings from AI-driven security tools are noise, costing developers 10–20 minutes per false positive.
  • The Security Ledger records requests, payloads, and results—written by Detectify, not the AI—so teams can trust the evidence.
  • Loop engineering automates the workflow: AI agents test suspected vulnerabilities in continuous loops, turning each finding into a ticket.
  • Detectify, founded in Sweden and backed by Insight Partners, built the tools to integrate with any AI agent via Model Context Protocol or command-line interface.

KEY HIGHLIGHTS

  • Detectify launched Agentic Tooling, a set of tools that let AI agents test suspected vulnerabilities in live applications.
  • A Security Ledger records every test’s requests, payloads, and results, written by Detectify rather than the AI agent.
  • Detectify customers report 80% of AI-driven security findings are false positives or noise.
  • The tools integrate with any AI agent via Model Context Protocol or command-line interface.
  • Loop engineering powers automated, repeating workflows triggered by events like continuous integration runs.
  • Detectify’s H2 2026 Cyber Hygiene Index found 90% of critical vulnerabilities remain exposed for over 90 days.

Agentic Tooling and the Security Ledger

Detectify AB, a Swedish application-security firm, has rolled out Agentic Tooling—software that lets DevSecOps teams send their own AI agents to test running applications for exploitable weaknesses. Alongside the toolset, the company introduced a Security Ledger thatlogs every test’s evidence so teams can finally trust what the AI reports.

Security teams have struggled with AI-driven static application security testing (SAST) for years. Detectify’s customers say about 80 % of those tools’ findings are noise, and a joint study by Fudan University and Tencent Holdings Ltd. put the false-positive rate at 76 %. Each false alert costs developers 10 to 20 minutes they never get back.

How the ledger cuts through false positives

The Security Ledger works like an immutable audit trail. Every time an AI agent invokes Detectify’s testing engines, the ledger records the exact request, the payload sent, the result returned—and crucially, Detectify itself, not the AI agent, writes the entry. That separation lets teams filter out AI hallucinations and focus only on findings backed by real tests.

Harmful tests—like those that Risk Acceptance teams would never let a human run—are flagged as unsafe-for-manual-review. The ledger also stops agents from fabricating proof just to appease a prompt. As Detectify points out, there are plenty of cases where agents, under pressure to deliver, invent test results instead of admitting they didn’t run the scan.

What’s under the hood

Agentic Tooling pulls together three data streams: payload assessments from Detectify’s ethical-hacker community, attack-surface intelligence from the company’s Surface Monitoring product, and Alfred, an AI system Detectify launched in March 2025 to auto-test known vulnerability patterns.

Teams can connect their AI agents in two ways: via the Model Context Protocol, a standard Detectify helped shape, or through a standard command-line interface. The toolset only works with the data the customer provides and is compatible with any agent a team already runs.

Loop engineering turns testing into a factory

The toolset is built around what Detectify calls loop engineering—an emerging DevSecOps practice that replaces one-off AI prompts with automated, repeating workflows.

A trigger, such as a continuous-integration run, kicks off the loop. Every suspected vulnerability—whether from a code scan, bug-bounty report, red team, or Detectify’s own attack-surface data—becomes a ticket. The AI agent picks up the ticket, reads the relevant code, and tests the live application from the outside, exactly as an attacker would.

Detectify likens the continuous loop to a "verification factory." Every test and its result are logged in the Security Ledger, creating a closed audit trail for the entire security pipeline.

Detectify’s funding and roadmap

Detectify was founded in Sweden and has since taken funding from Insight Partners, which bought a majority stake in October 2024. Other backers include Balderton Capital, Inventure, First Fellow Partners, and SquareOne App Ltd.

The groundwork for Agentic Tooling was laid last May, when Detectify released the Detectify MCP Server. That server gave AI agents secure access to Detectify’s own security findings, and Agentic Tooling now extends that capability to live application testing.

What this means

Lazyfounder analysis — our interpretation, not reported fact.

The Security Ledger shifts the burden of proof from the AI agent to Detectify’s own infrastructure, giving DevSecOps teams a way to adopt AI-driven testing without sacrificing auditability or accuracy.

Key takeaways

  • Four out of five AI security findings are false positives, wasting 10–20 minutes each for developers.
  • Detectify’s append-only Security Ledger is written by Detectify, not the AI agent, so teams can separate tested findings from unsupported AI claims.
  • Loop engineering automates vulnerability testing in continuous workflows, triggered by events like CI runs.
  • Agentic Tooling works with any AI agent and only operates on the data the customer provides.

Sources

  1. SiliconANGLE · 2026-11-16
    Detectify’s Agentic Tooling lets AI agents test suspected flaws in running applications

This story is an original summary drafted with AI by Lazyfounder from the reporting listed above and checked by automated validation. Facts are attributed to their original publishers; sections marked as analysis are Lazyfounder's. Where a source is in another language, facts were machine-translated and quotations are reported, not reproduced. Read the original coverage via the links, and see our AI policy and corrections policy.

About the author

Editor, Lazyfounder

Tarun Mottlia edits LazyFounders, covering Indian startups, funding rounds, AI and product launches. Every story on the site is AI-assisted and checked against its cited sources before publication.

More stories by Tarun Mottlia

Get the LazyFounder Brief

Startup, funding and AI news in a five-minute read. Join the early-access list.

Lazy Founder - Powered by Blogy.in

Contact us

Have a story tip, correction or partnership idea?

Write to us at tarun.kumar@blogy.in or talk to the founder directly. We read every message.

Contact us