TECHFUND says Hi AUDIT Agent hit 90.6% on CyberGym
TECHFUND said its Hi AUDIT Agent scored 90.6% on CyberGym, topping Anthropic’s published Claude Mythos 5 result by 6.8 points. The Tokyo-based accelerator says the security copilot is meant to bring vulnerability analysis and fix review into developers’ existing workflows.
Why it matters: - Hi AUDIT Agent is aimed at moving security review closer to the point where developers already work. - TECHFUND is positioning the tool as a way to help teams identify risks and review potential fixes without leaving their normal coding environment. - The CyberGym result gives TECHFUND a benchmarked performance claim in a crowded AI security market.
What happened: - TECHFUND announced that Hi AUDIT Agent achieved a 90.6% success rate on CyberGym. - The result surpassed Anthropic’s published Claude Mythos 5 score by 6.8 percentage points. - TECHFUND is a Tokyo-based technology accelerator. - The announcement was dated Oct. 15, 2026.
The details: - Hi AUDIT Agent is an AI-powered security copilot. - The product combines AI capabilities with specialized security knowledge and insights from security audits. - Developers can connect Hi AUDIT through the Model Context Protocol, or MCP, to compatible AI coding tools and editors. - Users can request security analysis, examine findings in context and review proposed fixes inside existing workflows. - TECHFUND said the product is designed to help development teams incorporate security reviews into daily work. - The company pointed readers to hiaudit.io for more information and free account creation. - CyberGym is an evaluation platform developed by a research team at the University of California, Berkeley. - CyberGym uses real-world software vulnerabilities. - The platform requires AI agents to generate working proof-of-concept tests that reproduce a specific vulnerability in a software project. - CyberGym’s reported success rate measures performance on those benchmark tasks.
Between the lines: - TECHFUND is framing Hi AUDIT Agent as infrastructure for security inside the development process, not as a standalone audit tool. - The CyberGym score suggests the product is being marketed on measurable capability, not just workflow convenience. - The comparison with Anthropic’s score is meant to signal competitive positioning in AI security tooling.
What's next: - Developers can create a free account and test the product through hiaudit.io. - TECHFUND is likely to use the CyberGym result to push adoption among development and security teams. - The company may lean on the benchmark score as it expands interest in Hi AUDIT and related security offerings.
The bottom line: - TECHFUND is using a strong benchmark result to argue that Hi AUDIT Agent can bring security analysis directly into the software development workflow.
Disclaimer: This article was produced by AGP Wire with the assistance of artificial intelligence based on original source content and has been refined to improve clarity, structure, and readability. This content is provided on an “as is” basis. While care has been taken in its preparation, it may contain inaccuracies or omissions, and readers should consult the original source and independently verify key information where appropriate. This content is for informational purposes only and does not constitute legal, financial, investment, or other professional advice.
Sign up for:
Technology Press Releases
The daily local news briefing you can trust. Every day. Subscribe now.
Check Your Email!
We sent a one-time activation link to: .
Confirm it's you by clicking the email link.
If the email is not in your inbox, check spam or try again.
Welcome back!
is already signed up. Check your inbox for updates.