Detectify, the Swedish application security platform built and trusted by hackers, today announced 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. A new Security Ledger lets security teams verify agents’ findings against objective evidence, ensuring informed decisions rather than relying solely on probabilistic conclusions.
The tools address a growing gap in DevSecOps. Engineering teams have built software factories that move code from commit to production in minutes. AI now writes a growing share of that code, and vulnerabilities move through the same pipelines at the same speed, which turns those software factories into vulnerability factories.
Detectify customers report that roughly 80% of the findings from their AI-driven static application security testing (SAST) tools are noise, showing that handing work to AI agents helps only if you can trust their output. Agents reason well, but they cut corners to deliver an answer and can produce a convincing report about a test they never ran.
Built for DevSecOps teams, Agentic Tooling gives agents access to the testing engines Detectify has built over more than a decade, including payload assessments sourced from its ethical hacker community, Alfred AI, and its Surface Monitoring data. The tools run in customers’ own environments through MCP and a CLI, use customers’ own agents and models, and operate only on the data customers provide.
Agentic Tooling is built for loop engineering, the emerging practice of running AI agents in automated, repeating workflows instead of prompting them by hand at every step. A CI run or scheduled job starts the loop. Findings from code scans, bug bounty reports, red teams, and Detectify’s attack surface data each become a ticket. An agent picks up the ticket, reads the relevant code, and uses Detectify’s tools to test the live application from the outside, as an attacker would, to confirm whether the issue can be exploited. Each pass takes one finding from trigger to proof, then moves on to the next. Run continuously, the loop becomes what Detectify calls a verification factory: the counterpart to the software factory, able to keep pace with the code it ships.
Detectify’s Security Ledger makes all of that work verifiable. Each time an agent uses Detectify’s tools, the ledger records the testing session, including the requests sent, payloads used, and results returned, and links the evidence back to the original ticket. Engineers can see what the agent did, what it concluded, and the proof supporting its assessment in a single timeline. Because the ledger is written by Detectify’s tools rather than the agent, and is append-only, teams can distinguish findings backed by actual tests from unsupported AI-generated conclusions. This allows organizations to scale security investigations without losing visibility into how decisions were reached.
The launch comes as security teams are already struggling to fix known vulnerabilities. Detectify’s H2 2026 Cyber Hygiene Index found that nine in ten open critical and high-severity vulnerabilities had stayed exposed for more than 90 days. Meanwhile, security teams face a growing volume of alerts that demand investigation. A 2026 study by researchers at Fudan University and Tencent found that 76% of the static-analysis alerts examined were false positives, and developers spent 10 to 20 minutes reviewing each.
Agentic Tooling builds on the Detectify MCP Server, released in May 2026, which gave AI agents access to Detectify’s security findings and scanning capabilities. Detectify plans to feed the loop with data from across its platform. For example, a newly opened port could automatically trigger an agent investigation.
Detectify Agentic Tooling is available on request starting November 16th, 2026. For more information, visit detectify.com
Related
This entry was posted on November 16, 2026 at 8:00 am and is filed under Commentary with tags Detectify. You can follow any responses to this entry through the RSS 2.0 feed.
You can leave a response, or trackback from your own site.
Detectify launches Agentic Tooling to bring verifiable evidence to AI-led security investigations
Detectify, the Swedish application security platform built and trusted by hackers, today announced 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. A new Security Ledger lets security teams verify agents’ findings against objective evidence, ensuring informed decisions rather than relying solely on probabilistic conclusions.
The tools address a growing gap in DevSecOps. Engineering teams have built software factories that move code from commit to production in minutes. AI now writes a growing share of that code, and vulnerabilities move through the same pipelines at the same speed, which turns those software factories into vulnerability factories.
Detectify customers report that roughly 80% of the findings from their AI-driven static application security testing (SAST) tools are noise, showing that handing work to AI agents helps only if you can trust their output. Agents reason well, but they cut corners to deliver an answer and can produce a convincing report about a test they never ran.
Built for DevSecOps teams, Agentic Tooling gives agents access to the testing engines Detectify has built over more than a decade, including payload assessments sourced from its ethical hacker community, Alfred AI, and its Surface Monitoring data. The tools run in customers’ own environments through MCP and a CLI, use customers’ own agents and models, and operate only on the data customers provide.
Agentic Tooling is built for loop engineering, the emerging practice of running AI agents in automated, repeating workflows instead of prompting them by hand at every step. A CI run or scheduled job starts the loop. Findings from code scans, bug bounty reports, red teams, and Detectify’s attack surface data each become a ticket. An agent picks up the ticket, reads the relevant code, and uses Detectify’s tools to test the live application from the outside, as an attacker would, to confirm whether the issue can be exploited. Each pass takes one finding from trigger to proof, then moves on to the next. Run continuously, the loop becomes what Detectify calls a verification factory: the counterpart to the software factory, able to keep pace with the code it ships.
Detectify’s Security Ledger makes all of that work verifiable. Each time an agent uses Detectify’s tools, the ledger records the testing session, including the requests sent, payloads used, and results returned, and links the evidence back to the original ticket. Engineers can see what the agent did, what it concluded, and the proof supporting its assessment in a single timeline. Because the ledger is written by Detectify’s tools rather than the agent, and is append-only, teams can distinguish findings backed by actual tests from unsupported AI-generated conclusions. This allows organizations to scale security investigations without losing visibility into how decisions were reached.
The launch comes as security teams are already struggling to fix known vulnerabilities. Detectify’s H2 2026 Cyber Hygiene Index found that nine in ten open critical and high-severity vulnerabilities had stayed exposed for more than 90 days. Meanwhile, security teams face a growing volume of alerts that demand investigation. A 2026 study by researchers at Fudan University and Tencent found that 76% of the static-analysis alerts examined were false positives, and developers spent 10 to 20 minutes reviewing each.
Agentic Tooling builds on the Detectify MCP Server, released in May 2026, which gave AI agents access to Detectify’s security findings and scanning capabilities. Detectify plans to feed the loop with data from across its platform. For example, a newly opened port could automatically trigger an agent investigation.
Detectify Agentic Tooling is available on request starting November 16th, 2026. For more information, visit detectify.com
Share this:
Like this:
Related
This entry was posted on November 16, 2026 at 8:00 am and is filed under Commentary with tags Detectify. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.