Reco Adds $55 Million as Investors Crowd Into AI-Agent Security
Reco has raised $55 million as it expands from securing software-as-a-service platforms toward tracking the AI agents operating inside them. TechCrunch reports that the financing extends a $30 million Series B announced in February and lifts the startup’s total capital raised to $140 million. AT&T, which is also a customer, invested through its venture arm; Forestay and Quadrille Capital also joined the extension.
The round arrives in a crowded field. TechCrunch says a quick look at public company profiles reveals at least two dozen firms selling some form of AI-agent security. Their tools take different approaches, but many promise to discover agents, monitor activity and govern what the agents can reach. Reco’s funding is a confirmed company announcement in the report. Whether its technology stands out from that competition is a different question, and the article supplies no comparative test of rival platforms.
Reco is moving from app maps to agent relationships
Until last year, Reco primarily sold tools to map and secure SaaS and AI platforms. Its current pitch uses a context graph linking agents with applications, people, accounts and permissions. In principle, that view would let a security team see the resources an agent can access and remove unnecessary privileges. The product’s claimed coverage matters because an agent may act through several systems, not just the application where it was created.
Co-founder and chief executive Ofer Klein told TechCrunch that companies are deploying agents faster than they can keep track of them. He said Reco discovered 21,000 agents at one Fortune 100 customer that the company had not known about. At a large financial-services customer, Reco claims it found an agent established by a former employee with access to Salesforce and an ability to share data with a domain outside the customer’s view. Both examples are vendor accounts from unnamed customers; TechCrunch does not independently verify the counts or describe the full circumstances of the latter access.
Reco says it integrates with more than 280 applications and can add new integrations within days. Klein also described using browser and network signals to find agents beyond directly connected apps and controls that inspect prompts and tool calls. These are descriptions of the platform’s design and claimed reach, not a guarantee that it discovers every hidden agent or prevents every unsafe action.
New money meets an unsettled market
The company’s investor list gives the round a customer connection through AT&T, but TechCrunch does not disclose how much each participant contributed or the financing’s terms. Klein said Reco’s valuation has more than doubled since the February round and placed it vaguely in the high hundreds of millions of dollars. He did not provide a precise figure. That estimate should not be presented as a verified transaction valuation.
Klein also said annual recurring revenue is currently in the double-digit millions of dollars and he expects it to triple this year. The present range is broad; the triple-growth statement is his forecast, not revenue already recorded. Reco says it has more than 100 customers, with financial-services firms contributing about 40% of its business. The report does not specify whether that share is measured by revenue or customer count.
TechCrunch includes claims from other vendors that illustrate why security teams are paying attention: HiddenLayer’s chief executive said more than 50 customers run agents touching critical systems or sensitive assets, while Cymphony said it found about 85,000 files at one company accessible to AI tools. These are separate companies’ examples, not independently comparable measurements of overall enterprise risk. They show how much the market’s evidence still depends on vendor-reported observations.
Financing does not settle the security question
Reco plans to use the new capital for hiring, sales, partnerships and customer support. That is a commercial expansion plan, rather than a reported deployment of a new defensive control. The company is betting that knowledge of existing SaaS applications gives it a useful vantage point as customers add agents and permissions multiply.
The hard test for buyers is whether that vantage point produces a reliable inventory and practical limits on access across systems. A large count of discovered agents may signal visibility, but it does not show which agents were harmful or whether controls prevented a breach. TechCrunch’s report establishes the financing and presents Reco’s approach and claimed customer examples; it does not provide an independent efficacy study. The $55 million round is therefore a clear sign of investor interest in agent security, while the protection delivered remains a question for customers to verify in their own environments.
