All news is verified and thoroughly reviewed by leading blockchain experts and seasoned industry insiders.
- CertiK has launched Skill Scanner, a security product designed to assess third-party AI skills before they are used by AI agents.
- The tool targets risks such as hidden malicious behavior, unauthorized data access, and autonomous execution in Web3 and Web2 environments.
CertiK is diving deeper into AI security with the launch of CertiK Skill Scanner, a product designed to review third-party AI skills before they are installed, deployed, or approved for use within enterprise systems.
The company describes the tool as something close to an antivirus layer for The AI was an agent. The comparison isn’t perfect, but it gets to the point. AI agents no longer just answer questions in a chat window. They began to invoke external tools, read files, run workflows, transfer data between systems, and in the most sensitive cases, interact with financial infrastructure.
This changes the security equation. A bad browser extension can actually be dangerous. Poor AI skill associated with an autonomous agent can be even worse, because the agent may perform actions quickly and in a context that is not fully visible to the user.
AI skills create new implementation risks
AI skills are becoming add-ons to the agent economy. It expands what an AI agent can do, from pulling data and automating tasks to executing financial actions and interacting with Web3 protocols. However, each additional skill also creates another point where something can go wrong.
This danger is not limited to obvious malware. A skill may request more data than it needs, behave differently during execution than it did during review, lead to unauthorized API calls, or create conditions for later misuse. In financial environments, anxiety becomes even more acute. A tool that can initiate finance calls, sign workflow requests, or set up transactions needs a different level of scrutiny than a simple productivity add-on.
CertiK said Skill Scanner is designed to detect hidden malicious behavior, unauthorized data access and execution risks before sensitive systems are exposed. Unlike broader AI-based scanning tools, the company says its product focuses on risks that can arise during actual execution, including situations involving money movements and financial transactions.
Rongwei Guo, CEO and co-founder of CertiK, said the security model around third-party skills is becoming more important as AI agents move into financial systems, enterprise workflows and everyday digital services.
“CertiK Skill Scanner is designed to create a unified trust layer prior to implementation, helping users and platforms identify hidden risks before exposing sensitive data, assets or systems,” said Gu.
Markets, institutions and developers are at the forefront
The first target groups are AI skills markets, enterprises and developers. Markets can integrate the scanner into their deployment pipelines, so skills are reviewed before deployment. They can also display CertiK judgments as trust indicators for users deciding whether to install a third-party skill.
For enterprises, the use case is more defensive. Companies testing AI agents internally need a way to evaluate third-party skills before they enter production environments or touch customer data, internal systems, or compliance-sensitive workflows. This is where the recorded review process becomes useful. It gives security teams something more concrete than a vendor claim or developer description.
Freelance developers can also use the scanner to self-skill audit before publishing. Future updates will expand direct access to regular users, allowing individuals to check skills themselves before installation or use, CertiK said.
The scanner produces a score from 0 to 100, plus judgments of ‘pass’, ‘warning’ or ‘fail’ and a limited list of scores grouped by severity. CertiK says the system reaches 90.5% accuracy in identifying security risks, with the aim of reducing false positives while making AI skills assessments more reliable.
The product is already deployed in selected Web3 AI agent infrastructure environments. CertiK is also working on integrations with additional AI Skill platforms, including FinChip.ai.
This launch follows CertiK’s broader expansion into AI-focused security infrastructure, after the company introduced its solutions AI Auditor initiative earlier this year. For a company known for its Web3 audits, this step is a logical extension. When AI agents begin handling code, assets, permissions, and business workflows, security checks should be performed before implementation, not after the system has already been exposed.
