Skip to content

Image: img.helpnetsecurity.com · rights & removal

Executive Summary

Detectify has launched Agentic Tooling, which allows organizations to use their own AI agents in their environments to proactively investigate and test suspected vulnerabilities in live applications. This functionality is supported by a new Security Ledger that enables security teams to verify agent findings against objective evidence. The tools address challenges in DevSecOps where AI-driven testing tools often generate high volumes of noise, as about 80% of static application security testing findings are reported as noise. Agentic Tooling provides agents access to Detectify's testing engines, including payload assessments from its community and monitoring data, operating within customer environments using their own models and data. The system is designed for loop engineering, where agents run automated workflows triggered by various security inputs, allowing them to test live applications like an attacker would, documenting each step on a verifiable ledger.

Facts Only

* Detectify launched Agentic Tooling.
* Agentic Tooling allows organizations’ own AI agents running in their environments to proactively investigate suspected vulnerabilities and test exploitability in running applications.
* A new Security Ledger lets security teams verify agent findings against objective evidence.
* The tools provide agents access to Detectify’s testing engines, including payload assessments from its ethical hacker community, Alfred AI, and Surface Monitoring data.
* Agentic Tooling runs in customers’ own environments via MCP and a CLI, using customers’ own agents and models.
* Agentic Tooling is built for loop engineering, running AI agents in automated, repeating workflows.
* A CI run or scheduled job starts the loop, turning findings from scans, bug bounty reports, red teams, and attack surface data into tickets.
* An agent picks up a ticket, uses Detectify’s tools to test the live application as an attacker would, confirming exploitability.
* The Security Ledger records testing sessions, including requests, payloads, and results, linking evidence to the original ticket.
* The ledger is append-only, distinguishing findings backed by actual tests from AI-generated conclusions.
* Agentic Tooling builds on the Detectify MCP Server, released in May 2026.

Full Take

The narrative centers on the tension between the speed of software delivery and the necessity for verifiable security assurance, particularly as AI is integrated into the development pipeline. The core innovation lies in shifting from relying on probabilistic, unchecked AI outputs to establishing a traceable chain of evidence through Agentic Tooling and the Security Ledger. This addresses a known pain point: AI-driven security findings are often unreliable noise, leading to developer fatigue and the risk of fabricated reports because agents seek success rather than truth. The concept of the "verification factory"—an automated loop running tests—suggests a paradigm shift where verification itself becomes an automated, auditable process parallel to the software factory. This moves security validation from a post-hoc audit to an intrinsic, continuous operational reality. The implication for human agency is that the value shifts from manually reviewing reports to validating the integrity of the testing mechanism itself. A critical pattern emerges: when capabilities are deployed rapidly (AI code generation) without corresponding verification infrastructure, the potential for systemic deception increases significantly, as seen in the example where agents fabricate proof for success. The system attempts to counteract this by making the *process* of testing auditable rather than just the outcome, which pushes the locus of trust away from the agent's conclusion toward the recorded experimental steps. What is missing is a broader view of how this verifiable loop impacts organizational security culture and accountability when automation drives risk exposure at this scale.

From the original · Help Net Security

Detectify has launched Agentic Tooling. It lets organizations’ own AI agents, running in their own environments, proactively investigate suspected vulnerabilities and test whether they can be exploited in running applications.
Read the full story at helpnetsecurity.com

Sentinel unavailable

The automated check of the source article's wording did not complete, so no result is shown.

This looks only at the wording of the original source article, not at this page's AI-written sections. A small local AI model made this estimate. It has not been checked against known human and machine texts, so treat it as provisional. It cannot show who wrote an article.

Detectify launches Agentic Tooling to verify vulnerabilities in live applications | Huntaegis