All analysis was generated autonomously, without human review. Scores are analytical opinions drawn from the cited public sources, without hands-on testing. They are not audits, certifications, investment reports, purchasing advice, or evaluations of quality.
Reco, founded in 2020, sells software that maps the AI agents, applications, identities, and permissions in large companies. It lets security teams see what each agent can reach and cut off access it doesn't need. Its CEO told TechCrunch that Reco has more than 100 customers and annual recurring revenue in the double-digit millions. He said financial services companies account for about 40% of its business. Named customers include AT&T, Nestle, Check Point, and Tampa General Hospital. BankInfoSecurity reports Reco employs 189 people. Reco has raised $140 million, including a $55 million round led by AT&T Ventures. Reco says it covers agents on every vendor's platform. If Microsoft or Salesforce adds controls for agents inside its own software, fewer buyers would need a separate product.
| Description | Reco is a security company whose platform maps the AI agents, SaaS applications, identities, and permissions in an enterprise, shows what each agent can reach, and governs or blocks risky agent access and actions. | [f1] |
|---|---|---|
| Founded | 2020 | [f2] |
| HQ | New York, New York, United States | [f3] |
| Funding | $140M total | [f4] |
| Latest funding | $55M strategic round, September 2026, with AT&T Ventures, Forestay, and Quadrille Capital | [f3] |
| Product | What it does |
|---|---|
| Reco Platform | Maps agents, apps, identities, and permissions in the Reco Graph, with Agent Governance, Agent Security, and Agent Runtime layers for inventory, access control, and runtime policy. |
| Reco Identity Threat Detection and Response | Detects compromised identities and suspicious behavior in SaaS environments in real time, flagging identity-based threats across connected applications. |
| Reco Shadow AI Discovery | Discovers shadow AI tools, embedded AI features, and AI agents across SaaS apps, maps each agent to an owner and risk score, and governs generative AI usage across the enterprise. |
Cyber Defense Matrix
| Identify | Protect | Detect | Respond | Recover | |
|---|---|---|---|---|---|
| Devices Workstations, servers, phones, tablets, storage, network devices, IoT infrastructure, and similar hardware. | |||||
| Applications Software, interactions, and application flows on the devices. | |||||
| Networks Connections and traffic flowing among devices and apps, plus communication paths. | |||||
| Data Content at rest, in transit, or in use across devices, apps, and networks. | |||||
| Users The people using the devices, apps, networks, and data. |
The Reco Platform maps SaaS applications and the identities and permissions tied to them, and Reco Identity Threat Detection and Response surfaces compromised identities and risky behavior. These capabilities are mapped to the Cyber Defense Matrix. [f5]
AI Defense Matrix
| Govern | Identify | Protect | Detect | Respond | Recover | |
|---|---|---|---|---|---|---|
| AI-Workload Platforms Inference servers, training platforms, vector DB platforms, and the model-loading supply chain. | ||||||
| AI Coding and Orchestration Tools AI coding tools and agentic orchestration tools on user devices, plus their plugins, skills, hooks, system prompts, scaffolding, harnesses, configuration settings, and MCP clients. | ||||||
| AI-Generated Code Code produced by AI tools, AI-assisted reviews, AI-generated infrastructure-as-code and tests, and vibe-coded apps that bypass CI/CD. | ||||||
| AI Traffic Paths to external and self-hosted AI services, MCP tool channels, agent-to-agent calls, model-registry downloads, and egress to unapproved AI services. AI gateways, LLM routers, and MCP gateways steer traffic along those paths. The steering decisions and their inputs belong here too: routing policies, MCP server registries, and agent naming and discovery services. | ||||||
| AI Model Model weights, fine-tuning checkpoints, model cards, registries, an AI bill of materials (AIBOM), and the third-party LLMs your enterprise consumes. | ||||||
| Training Data Datasets used for training, fine-tuning, and continued learning. | ||||||
| Runtime AI Data User prompts, inference inputs, RAG content, vector DB content, persistent agent memory, and interaction history. | ||||||
| AI Agent Identities AI agents as non-human principals, plus credentials, keys, permission scopes, service accounts, and delegation chains across agents and tools. |
Reco Shadow AI Discovery finds shadow AI tools and embedded AI features across SaaS apps, maps each AI agent to an owner and a risk score, and tracks data movement to unauthorized AI systems. These capabilities are mapped to the AI Defense Matrix. [f6]
How well the company can compete in its security market, scored across eight dimensions against public evidence.
| Dimension | Score | Rationale |
|---|---|---|
| Problem Clarity How precisely the company defines its problem, with evidence the problem exists at the scale claimed. | 3/5 | Reco names its buyer, the enterprise security team, and its pain: AI agents that arrive inside applications with no review or owner. The quantified pain is Reco's own, such as the 21,000 unknown agents its CEO told TechCrunch the platform found at one Fortune 100 customer, so the problem is clear but not independently quantified. [s1, s4, s15] |
| Capability Depth How specific the technical capabilities are, with evidence beyond marketing claims such as docs and third-party validation. | 3/5 | Reco's product pages describe concrete mechanisms: a graph linking agents, apps, identities, and permissions, runtime policies that block, warn, redact, or allow, and alerts sent to SIEM and SOAR tools. The outside voices in the reviewed sources are customer quotes on Reco's own site and press relaying the company's account, so no third-party evaluation of these capabilities appears. [s2, s3, s6, s1, s15] |
| Market Timing Whether the market is ready for this product, with evidence that buyers are actively seeking solutions. | 3/5 | Reco's CEO told TechCrunch in September 2026 that companies were deploying agents faster than they could track them over the preceding year. The buyer-side evidence is one kind, named customer relationships such as AT&T's, which TechCrunch reports, while its count of at least two dozen agent-security vendors measures supply rather than buyer demand. [s15, s14] |
| Team Credibility Demonstrated domain expertise with public signals such as prior exits, publications, and industry recognition. | 3/5 | The founders bring verifiable relevant experience: Tal Shapira holds a Ph.D. and formerly headed a cybersecurity R&D group in the Israeli Prime Minister's Office, where Gal Nakash served as a lieutenant colonel. ISMG reports that Ofer Klein spent nearly eight years leading the IoT platform kwik.me, and the reviewed sources name no prior security product build or exit for the founders. [s8, s17] |
| GTM Proof Evidence of actual traction (customers, revenue signals, partnerships) beyond stated intentions. | 4/5 | Named customers come from several independent and company sources: SiliconANGLE lists Check Point, Cato Networks, Desjardins Group, Nestle, and UiPath, TechCrunch reports AT&T as a customer, and Reco publishes stories from Tampa General Hospital and Banco BS2. The scale figures, more than 100 customers and recurring revenue in the double-digit millions, are the CEO's own as relayed by TechCrunch, so customer scale is not independently corroborated. [s19, s15, s10, s14] |
| Funding Efficiency Whether funding matches go-to-market ambition, with signs of capital-efficient growth. | 3/5 | Reco has raised $140 million in total, adding a $55 million round in September 2026 after a $30 million Series B in February, and it shipped agent-security and runtime products. Revenue appears only as the CEO's double-digit-millions band, so the raises look proportional to an enterprise motion while output per dollar stays unconfirmed. [s15, s16, s18, s24, s21] |
| Category Clarity Whether the company creates or fits a recognizable category that buyers can quickly place in their stack. | 3/5 | Reco fits a recognizable category: TechCrunch, SecurityWeek and ISMG all describe it in agent security terms. TechCrunch also reports that Reco mostly sold software to map and secure SaaS and AI platforms before it repositioned, and that rival promises to discover and govern agents sound quite similar. Reco shows its GigaOm SSPM Radar Leader placement on its own blog and newsroom, so placement in the current category is recent and contested rather than settled by analysts. [s15, s16, s17, s23] |
| Incumbent Defensibility How vulnerable the core value proposition is to absorption as a feature by a platform vendor. | 3/5 | Replicating Reco's coverage would take more than one feature release from an adjacent platform: Reco reports more than 280 app integrations and 1,000 detection controls feeding one graph, and AT&T's CISO credits it with governance across the company's enterprise apps and AI. That is friction rather than a structural moat, because the reviewed sources show no data asset or procurement position that a platform vendor could not eventually build or buy. [s14, s2, s15] |
Reco sells to enterprise security teams that need to know which AI agents are running in their company and what each one can reach. Reco's own pitch is that agents often arrive switched on inside applications the company already runs, with no procurement, review, or owner.
Press described Reco as an AI SaaS security company whose growth followed the rise of shadow AI. TechCrunch reported that Reco had repositioned around a context graph that connects agents to apps, people, accounts, and permissions.
Reco's figures show the scale it claims. Its CEO told TechCrunch the platform found 21,000 unknown agents at one Fortune 100 customer. ISMG reports that agent governance is now Reco's most common use case, meaning discovering, assessing, and approving or rejecting agents. [s1, s15, s17, s18, s20]
The Reco Platform has three layers built on one graph. Agent Governance inventories agents, apps, identities, and connections. Agent Security assigns each agent an owner, a scope, and a risk score and flags permissions wider than the job requires. Agent Runtime applies policies that block, warn, redact, allow, or log agent actions.
Reco builds the graph from its integration catalog. Reco says the Reco Factory builds new integrations in hours and feeds the Reco Library, and its September 2026 announcement counts more than 280 app integrations and 1,000 detection controls. Earlier lines remain on the site, including identity threat detection that sends alerts to SIEM and SOAR tools and shadow AI discovery.
Reco was still completing the runtime layer in mid-2026. Reco put browser-based discovery in early access and targeted October for general availability of response and tool-call analysis and blocking. The reviewed sources describe these capabilities in Reco's words, and no independent technical evaluation appears. [s2, s3, s4, s5, s6, s7, s14, s21, s22]
Reco competes in a crowded field. TechCrunch counted at least two dozen companies selling some form of AI agent security in September 2026 and wrote that their promises to discover and govern agents sound similar.
Reco argues that it stands apart through breadth across platforms. Its CEO told TechCrunch the platform finds agents outside the apps it connects to directly by using browser and network signals, and Reco's pages describe one graph spanning every platform where agents appear.
Reco also holds an analyst placement from its earlier market. It announced a Leader and Fast Mover placement in GigaOm's 2025 Radar for SaaS security posture management, and the reviewed sources carry that placement only through Reco's own announcement. GigaOm's 2025 report covers SaaS security posture management. [s15, s2, s4, s23, s13]
Reco's traction is now visible through both named customers and company-reported scale. SiliconANGLE named Check Point, Cato Networks, Desjardins Group, Nestle, and UiPath among its customers in February 2026. TechCrunch reported AT&T as a customer that also invested in September 2026.
Reco's own pages add named stories. They include Tampa General Hospital, Banco BS2, and BigID, and AT&T's chief information security officer is quoted in Reco's funding announcement on the governance Reco provides.
The scale figures come from the company. Reco's CEO told TechCrunch that Reco has more than 100 customers, that financial services make up about 40 percent of the business, and that annual recurring revenue is in the double-digit millions and expected to triple this year. ISMG reports, also from the CEO, that multiple Fortune 500 customers run the agent security product in production. [s19, s15, s10, s14, s17]
Reco's three founders still run the company. Ofer Klein is chief executive, Gal Nakash is chief product officer, and Tal Shapira is chief technology officer.
Tal Shapira has a security research background. Tal Shapira holds a Ph.D. from Tel Aviv University, formerly headed a cybersecurity R&D group in the Israeli Prime Minister's Office, and sits on a Cloud Security Alliance working group on AI controls. Gal Nakash is a former lieutenant colonel in the same office.
Ofer Klein's prior company was an IoT platform. ISMG reports that he spent nearly eight years leading the IoT platform kwik.me and nearly eight years at Elbit Systems. Calcalist counted about 100 employees, and ISMG counted 189. [s8, s17, s20]
Reco publishes the attestations enterprise buyers check during procurement. Its trust center lists SOC 2 Type 2, ISO/IEC 27001:2022, CSA STAR Level 1, and ISO/IEC 42001:2023, the management-system standard for AI.
The trust center also answers a common residency question. It lists United States or Europe hosting based on customer selection. [s12]
| Company | Relationship | Note | Compare |
|---|---|---|---|
| Obsidian Security | competes with | Competes for the same enterprise buyer of SaaS security posture management and AI agent governance. | |
| Grip Security | competes with | Competes for the same enterprise buyer of SaaS discovery and shadow AI governance. | |
| Valence Security | competes with | Competes for the same enterprise buyer of SaaS security posture management. | N/AWe captured the evidence for these companies under different evidence-model versions (v1 vs v2), so the totals were scored under different conditions and are not directly comparable. |
Add analyzed competitors to compare them side by side with Reco.
A closer look at the company's product strategy, measuring how defensible it is against market forces and examining the eight areas behind it.
reinforce or reposition
Reco sells software that enterprise security teams run themselves, and customers who adopt it build integration work around it. Leaving would mean reconnecting applications, rebuilding each agent's owner and policy assignments, and rewiring alerts into the team's other security tools. Reco's named buyers are large enterprises, many in banking, telecommunications, and healthcare. Its certifications, including an AI management standard, help it pass enterprise security reviews. The asset Reco names is its Reco Library, a catalog of more than 280 security-built integrations.
| Dimension | Score | Rationale |
|---|---|---|
| Value Delivery Does the product sell software as the product, or judgment, trust, or accountability with software as the delivery mechanism. | 1/3 | Reco sells software that the customer's security team runs and whose outcomes it owns. The service layer in the reviewed sources belongs to partners, such as the managed IT provider that offers MDR services on Reco's threat detection. |
| Switching Cost How expensive leaving is for a customer: data portability, integrations, learned workflows, network effects, regulatory data residency. | 2/3 | Leaving Reco means reconnecting applications, rebuilding each agent's owner and policy assignments, and rewiring the alerts Reco sends into SIEM and SOAR tools, which is meaningful integration and workflow friction. This switching work is an inference from the documented integrations. The cited record does not size the migration, and it shows no network effect, non-portable state, or regulatory constraint tied to Reco. |
| Compliance Moat Whether certifications, liability acceptance, or audit trails block an easy replacement. | 1/3 | Reco's trust center lists SOC 2 Type 2, ISO/IEC 27001:2022, ISO/IEC 42001:2023, and CSA STAR Level 1, attestations a funded rival can obtain through ordinary preparation. The probed trust center shows no government authorization that would make replacing Reco slow. |
| Problem Complexity Whether the product requires ML, optimization, real-time systems, or years of specialized expertise. | 3/3 | Reco correlates agents, apps, identities, and permissions from more than 280 integrations into one graph and inspects prompts and tool calls in real time. That combination of graph analysis and real-time enforcement takes years of specialized expertise to build. |
| Buyer Profile Whether buyers are SMB operators, mid-market IT teams, or regulated enterprises and governments with procurement gates. | 3/3 | Reco is built and sold for large regulated enterprises: its CEO told TechCrunch financial services make up about 40 percent of the business, and the named buyers include AT&T, Desjardins Group, Banco BS2, and Tampa General Hospital. The 40 percent share is the company's own figure, while the named customers come from press and Reco's customer stories. |
| Layer Whether the product is an end-user application, a platform with application features, or infrastructure other applications depend on. | 2/3 | Reco's governance and security layers sit beside the customer's applications with read-only sensors and no proxy, a platform with application features. Its runtime layer can block actions through app-native hooks and endpoint controls. Reco targeted early access for browser blocking and October general availability for response and tool-call analysis and blocking, and the reviewed sources show no deployment in the path of production traffic. |
| Proprietary Data, Content, or IP Whether the product accumulates datasets, content licenses, or IP that a rival cannot recreate from scratch. | 1/3 | The retained asset Reco names is the Reco Library, a catalog of security-built integrations, which is software Reco ships rather than a dataset, content licence, or patent. Reco says it analyzes billions of monthly interactions, but the reviewed sources describe the graph as a map of each customer's environment and show no data accumulated across customers. |
Reco sells to large enterprises, and financial services is a large part of its business. Reco's CEO told TechCrunch that financial services make up about 40 percent of the business across more than 100 customers. SiliconANGLE described the customer base in February 2026 as multiple Fortune 500 and global enterprises across financial services, healthcare, technology, pharma, and manufacturing.
The named buyers fit that description. They include AT&T, Desjardins Group, Banco BS2, Tampa General Hospital, Nestle, Check Point, and Cato Networks. ISMG reports the CEO's view that big enterprises are adopting agents early, and those are the customers Reco already serves.
Reco's choice of segment follows a timing claim. Reco describes agents inside applications enterprises already run. Its CEO told TechCrunch that Reco broadened its scope because companies were deploying agents faster than they could track.
The Reco Graph is the core of the product. Reco says it maps relationships among agents, identities, applications, permissions, data, and workflows, and SecurityWeek describes it mapping agents to accounts, API connections, sessions, MCP tools, and workflows. Agent Security uses that map to assign owners, scopes, and risk scores and to flag excess permissions.
Reco applies AI to the graph itself. Its CEO told ISMG that Reco uses its own AI and agentic technology to analyze the graph and decide on actions across billions of monthly interactions. These are the company's descriptions, and no independent evaluation of detection quality appears in the reviewed sources.
Integration speed is the claim a buyer can test. Reco says its no-code Factory adds applications in days, and its announcement counts more than 280 integrations, each built to pull identity, permission, and activity data.
Reco runs an enterprise sales motion. Its site routes buyers to demo requests, and the September 2026 round is earmarked for sales, partnerships, channels, and customer support, according to SecurityWeek.
Reco is adding partners to reach buyers. Its newsroom lists partnerships with ServiceNow and SecurityScorecard. Reco's customers page also shows a managed IT provider that delivers MDR services on Reco's threat detection.
AT&T is both an investor and a customer reference. AT&T invested through its venture arm and is a customer, TechCrunch reports, and AT&T's CISO is quoted in Reco's announcement. ISMG reports the CEO's plan to partner with large SASE vendors to add an agent security layer.
Reco's pricing URL returns the site's not-found page, and Reco routes buyers to demo requests. The reviewed sources do not say whether Reco charges by user, application, or agent, so a buyer has to learn the unit during evaluation.
Its governance page says there is no proxy to deploy and no traffic to reroute, and that seven sensors with read-only access feed the platform.
The product feeds existing security operations. Reco's threat detection sends alerts to SIEM and SOAR platforms, and its governance layer produces audit reports aligned to the EU AI Act, ISO 42001, SOC 2, and NIST AI RMF.
Reco describes runtime enforcement through app-native hooks and endpoint controls. Its runtime page says that enforcement can use the customer's gateway or Reco's. The trust center lists United States or Europe hosting at customer selection.
Reco's trust center lists the attestations an enterprise security review expects. They include SOC 2 Type 2, ISO/IEC 27001:2022, and CSA STAR Level 1.
Reco also lists AI-specific governance paperwork. The trust center includes ISO/IEC 42001:2023, the AI management-system standard, which matches what an AI governance buyer will ask a vendor that oversees agents.
Reco presents a three-layer platform on one graph. Its governance, security, and runtime layers draw on the Reco Graph and the Reco Library, which supply the platform's app, agent, and relationship context.
Reco depends on breadth across other vendors' platforms. The platforms that host agents, such as Microsoft 365 and Salesforce, could add their own oversight. Reco's answer, in its own words, is that other vendors only see agents inside their own ecosystems while Reco covers every platform where agents appear.
Reco is adding partners around the platform. It lists partnerships with ServiceNow and SecurityScorecard, and its CEO told ISMG that Reco plans alliances with SASE vendors and agent platform makers.
The three founders hold the chief executive, product, and technology roles. Tal Shapira, the technology chief, holds a Ph.D. and formerly headed a cybersecurity R&D group in the Israeli Prime Minister's Office. Gal Nakash, the product chief, is a former lieutenant colonel in the same office.
Ofer Klein, the chief executive, came to Reco after leading an IoT platform. ISMG reports nearly eight years leading the IoT platform kwik.me and nearly eight years at Elbit Systems.
Calcalist counted about 100 employees, split between Israel and the United States. ISMG counted 189.
| Id | Source | Tier | Accessed |
|---|---|---|---|
| f1 | https://www.reco.ai/about-us | official | 2026-10-16 |
| f2 | https://siliconangle.com/2026/02/10/reco-raises-30m-lock-ai-apps-autonomous-agents/ | press | 2026-10-16 |
| f3 | https://www.securityweek.com/reco-raises-55-million-for-agentic-security/ | press | 2026-10-16 |
| f4 | https://techcrunch.com/2026/09/29/reco-raises-55m-as-ai-agent-security-startups-crowd-the-market/ | press | 2026-10-16 |
| f5 | https://www.reco.ai/ | official | 2026-10-16 |
| f6 | AI Defense Matrix Catalog mapping | other | 2026-10-03 |
This site is an experimental research aid created by Zeltser Security Corp. All its data gathering and analysis was performed autonomously without human review, and it can contain errors of fact, interpretation, and judgment that a human reviewer might catch.
The analyses are statements of opinion, not statements of fact. Machine analysis produced the scores, summaries, and matrix placements by weighing the public sources each page cites, and reasonable people can weigh the same sources differently. Where a page states a fact, it cites the public source and the date it was checked, and the statement is only as accurate as that source. Unless a profile expressly says otherwise, the analysis involves no hands-on testing and no independent validation of any company's products or services.
Nothing here is professional, security, legal, financial, investment, or purchasing advice, and nothing here is a recommendation to invest in, do business with, or avoid any company. Inclusion of a company is not an endorsement, and absence of a company is not a judgment about it. Reading this site creates no advisory or client relationship. Verify any detail you plan to act on against the vendor's current materials.
The content is provided "as is" and "as available," with all warranties disclaimed, express or implied, including merchantability, fitness for a particular purpose, accuracy, and non-infringement. No entry is warranted to be complete, current, or correct. Companies change, vendors update their claims, sources can be wrong, and automated analysis can misread them.
To the fullest extent permitted by law, the operator, Zeltser Security Corp, is not liable for any damages that arise from using this site or relying on its content, including direct, indirect, incidental, special, and consequential damages and lost profits, even if advised that such damages were possible. If you are dissatisfied with the site or disagree with these terms, your remedy is to stop using it.
Entries link to vendor pages, press coverage, and other external sites that Zeltser Security Corp does not control and is not responsible for. A link is not an affiliation with the destination or an endorsement of it. Product and company names and trademarks are the property of their owners, used here nominatively to identify the companies described. Short quotations from cited sources appear for identification and commentary.
Use, quotation, automated retrieval, and redistribution of the content are governed by the Terms of Use at cybercompanyprofiles.com/terms, which permit personal and internal business use with attribution and prohibit republication and resale.