Creators of Hit, the enterprise-friendly, open-source variant of OpenClaw NanoClaw partner with a leader in software supply chain management JFrog Launching a new, collaborative security integration that will protect NanoClaw autonomous agents from malicious code injection.
"These agents do things that you can’t necessarily control and you can’t necessarily train." JFrog Chief Strategy Officer Gal Marder told VentureBeat in an exclusive interview.
Available immediately, the partnership connects NanoClaw agents directly to JFrog’s audited software registries, ensuring AI assistants can pull only scanned, safe dependencies.
The release addresses a rapidly growing blind spot in the technology: autonomous agents often install packages in the background to extend their capabilities, often without the knowledge or oversight of human operators.
"The people running the agents aren’t necessarily developers and aren’t even aware of their implications." Gavriel Cohen, founder and CEO of NanoClaw and co-founder of the new commercial services startup, explained, NanoCo AI.
There are partners to enable a wider ecosystem trying to do is completely free to the open source community, while enterprise organizations can seamlessly deploy their agents through existing, commercially licensed JFrog environments.
The new technical capability provided by this partnership follows NanoCo’s actions to add permission dialogs between the applications it has available through. Collaboration with Verceland a New collaboration with Docker to allow NanoClaw agents to run more securely in isolation from other application environments directly in Docker virtual containers.
The risk of current, personal autonomous AI agents
When an operator interacts with an autonomous system like NanoCo’s NanoClaw, it communicates at a high level of abstraction.
The user can simply send an audio file or voice recording to invite the agent to independently determine how to process it.
As Cohen explains, the agent thinks that "oh i can’t understand voice notes so let me go get the package and something discount install install install and run".
This dynamic self-improvement makes AI agents incredibly powerful, but also makes them highly vulnerable to chainsaw attacks.
Bad actors are increasingly poisoning open source registries with malicious packages. Agents bypass human control because they act autonomously to get what they need.
Operators, who may not even be developers, are largely unaware of the security implications that occur behind the scenes.
How NanoCo and JFrog work to stop agents from running malicious code
The integration between NanoCo and JFrog acts as an automated immune system for these AI environments.
Under the hood, NanoClaw agents are now configured to route their requests for software packages, CLI tools, and Model Context Protocol (MCP) servers only through JFrog registries.
If an agent tries to download a compromised library, such as a vulnerable version of the popular Axios package, the JFrog registry intercepts the request.
Blocks the installation, returning a security policy error to the agent, noting that the request was made "403 denied by JFrog registry with security policy".
Importantly, the system does more than just prevent danger; creates a dynamic correction loop. The agent is notified of the vulnerability and instructed to automatically search for and install a verified, non-malicious version of the required package instead.
For large organizations, this integration solves the headache of mass compliance. Marder notes that as enterprises adopt autonomous agents, they will definitely require visibility.
Organizations need it "we need a logging system, a place to track which agents are running by whom and which packages they are consuming and which skills they are using and which MCPs they are using." he told VentureBeat.
In addition to visibility, JFrog integration provides a foundation "trust layer" and strict management of what is allowed into these automated systems.
Licensing and Availability
In software distribution, licensing and access parameters dictate adoption. The NanoCo and JFrog partnership uses a two-pronged approach to serve both individual open source developers and highly regulated enterprises.
The integration is completely free for the open source community. JFrog provides open source NanoClaw users with free access to secure, vetted artifacts, tools, and skill resources.
This allows individual developers to manage native autonomous agents without being bogged down in manual approval requests for each dependency. In addition, community members build and share new ones "skills" for agents, these contributions are uploaded to the registry, scanned for malicious code, and cleaned before anyone can use them.
This infrastructure directly neutralizes the threat of poisoned community repositories.
The architecture for enterprise deployments seamlessly integrates into an organization’s existing commercial environment. Instead of using the public open source registry, corporate users point NanoClaw agents to their own internal JFrog registries.
This ensures that all agent activity complies with the company’s specific commercial licenses, internal security policies, visibility needs and governance standards.
As AI continues to blur the line between human intent and machine execution, the infrastructure that enables that execution must evolve. This partnership recognizes a basic reality: you can’t train an AI to perfectly recognize every zero-day vulnerability; Instead, you should create an environment where the agent cannot reach the vulnerability in the first place.
