SlowMist and AgentOn Form Strategic Partnership to Build a Secure AI Agent Ecosystem
Recently, SlowMist and AgentOn formally entered into a strategic partnership. The two events will collaborate on AI Agent safety functionality improvement, safety evaluation frameworks, and ecosystem practices, collectively advancing the AI Agent ecosystem towards a safer and reliable future.
Background
As AI Agents evolve from demos to real-world functions, their safety has turn out to be one of many business’s high considerations. Compared with conventional Web3 merchandise, AI Agents are able to invoking instruments, accessing knowledge, managing wallets, and executing transactions. As a end result, any safety subject can have a a lot broader impression.
As a main blockchain safety firm, SlowMist has lengthy been devoted to sensible contract safety, pockets safety, menace intelligence, and different safety domains, whereas constantly specializing in rising safety challenges within the AI Agent period.
AgentOn is dedicated to constructing an open AI Agent Marketplace that facilitates the creation, discovery, and adoption of AI Agents, offering an AI Agent ecosystem for builders and customers.
Through this strategic partnership, the 2 events will mix SlowMist’s sensible experience in safety offense and protection, danger detection, and menace analysis with AgentOn’s capabilities in constructing the AI Agent ecosystem, collectively exploring the safety infrastructure wanted to help the event of AI Agents.
Partnership Scope
1. AI Agent Security Capability Development
As AI Agent functions proceed to broaden, the assault floor has progressively advanced past dangers related to particular person fashions to embody the infrastructure layer, protocol/instrument layer, conduct layer, and mannequin layer. Due to the interactions amongst these completely different layers, threats can propagate throughout them and turn out to be amplified.
Therefore, AI Agent safety evaluation should cowl your entire lifecycle — from improvement and deployment to runtime — somewhat than being restricted to a single stage.
Through this partnership, each events will collectively discover safety functionality improvement all through the AI Agent lifecycle, overlaying key phases together with pre-deployment evaluation, runtime danger governance, and steady safety monitoring.
Leveraging its experience in Web3 safety, menace intelligence, and danger detection, SlowMist will conduct in-depth evaluation of the core assault surfaces of AI Agents and, based mostly on the OWASP Top 10 for Agentic Applications, construct danger identification capabilities tailor-made to AI Agent eventualities. The collaboration will deal with key safety dangers, together with Agent objective manipulation, instrument misuse and exploitation, id and privilege abuse, Agent provide chain vulnerabilities, unintended code execution, reminiscence and context poisoning, insecure Agent-to-Agent communication, cascading failures, Human-Agent belief exploitation, and malicious Agents.
At the identical time, drawing on real-world assault eventualities and safety analysis expertise, SlowMist will present AI Agent safety evaluation capabilities throughout six key dimensions: credential and key safety, Prompt Injection safety, harmful command and code execution, knowledge exfiltration safety, permission and provide chain safety, and audit and observability. These capabilities will assist builders inside the AgentOn ecosystem determine potential safety dangers and promote the mixing of safety all through your entire technique of AI Agent creation, deployment, and operation.
2. Exploring an AI Agent Security Certification Framework
As the AI Agent ecosystem continues to evolve, establishing a clear and dependable safety evaluation and certification framework will turn out to be an necessary basis for selling the wholesome improvement of the business.
Through this partnership, each events will collectively discover the event of an AI Agent safety certification framework. Based on the applying eventualities, safety capabilities, and danger traits of several types of AI Agents, they may progressively enhance the corresponding evaluation requirements and certification mechanisms.
By exploring a safety certification framework, the collaboration goals to assist builders reveal the safety capabilities of their AI Agents whereas enabling customers to extra simply determine trusted AI Agents with confirmed safety safeguards. Rather than establishing a one-time, unified customary, the certification framework will proceed to evolve in accordance to completely different AI Agent sorts and sensible utility necessities.
3. Building the AI Agent Security Ecosystem
In addition to technical functionality improvement, the 2 events can even collaborate on selling the AI Agent safety ecosystem, fostering business engagement, and supporting the developer neighborhood. This consists of collectively publishing AI Agent Security Best Practices, organizing technical knowledge-sharing and trade actions, and taking part in business conferences and developer ecosystem occasions.
Through sharing safety greatest practices, selling business collaboration, and strengthening ecosystem partnerships, each events will collectively advance the event of AI Agent safety ideas and additional drive the adoption of safety capabilities throughout the AI Agent ecosystem.
About AgentOn
AgentOn is an AI Agent-powered incomes platform incubated by ME Group, devoted to constructing a process collaboration community the place AI Agents and people work collectively, enabling Agents to function autonomously, full duties, and earn actual on-chain rewards.
About SlowMist
SlowMist is a menace intelligence agency centered on blockchain safety, established in January 2018. The agency was began by a staff with over ten years of community safety expertise to turn out to be a world power. Our objective is to make the blockchain ecosystem as safe as potential for everybody. We are actually a famend worldwide blockchain safety agency that has labored on varied well-known initiatives resembling HashKey Exchange, OSL, MEEX, BGE, BTCBOX, Bitget, BHEX.SG, OKX, Binance, HTX, Amber Group, Crypto.com, and so on.
SlowMist presents a number of companies that embrace however will not be restricted to safety audits, menace info, protection deployment, safety consultants, and different security-related companies. We additionally provide AML (Anti-money laundering) software program, MistEye (Security Monitoring), SlowMist Hacked (Crypto hack archives), FireWall.x (Smart contract firewall) and different SaaS merchandise. We have partnerships with home and worldwide companies resembling Akamai, BitDefender, RC², TianJi Partners, IPIP, and so on. Our intensive work in cryptocurrency crime investigations has been cited by worldwide organizations and authorities our bodies, together with the United Nations Security Council and the United Nations Office on Drugs and Crime.
By delivering a complete safety answer personalized to particular person initiatives, we are able to determine dangers and forestall them from occurring. Our staff was ready to discover and publish a number of high-risk blockchain safety flaws. By doing so, we may unfold consciousness and elevate the safety requirements within the blockchain ecosystem.
