MikeTrendsTrends right now
AI: reposAPI

search

AI security tools

Trends

  1. 1
    Pi pod lets developers run coding agents in self-hosted sandboxes●Show HN: Pi pod – Run your pi coding agent in sandboxes on your own serverYhnScienceBiology1216 min ago

    A developer has released Pi pod, a tool for running the Pi coding agent in isolated sandboxes on your own server. The launch lets programmers give AI coding assistants a controlled environment on self-hosted infrastructure instead of third-party clouds, addressing privacy and cost concerns. It was shared on Hacker News, where it drew immediate attention from developers interested in self-hosting AI coding tools.

  2. 2
    Cloudflare releases open-source security audit tool for coding agents●cloudflare/security-audit-skillβ¬’github5764 min ago

    Cloudflare has published security-audit-skill, an open-source JavaScript project that gives coding agents a framework for running multi-phase security audits. The tool produces machine-readable findings that are independently verified, aiming to make automated code reviews more trustworthy. Developers are sharing and star-ring the repository as interest grows in extending AI coding assistants beyond writing code into security testing.

  3. 3
    Security flaw in MCP protocol puts AI agents at risk●MCP for agent-to-agent comms may be the riskiest protocol you've never heard ofYhnTechnologyCybersecurity641 min ago

    A newly reported vulnerability affecting AI agents from Google and other companies highlights a structural weakness in the Model Context Protocol, the emerging standard that lets AI agents communicate with tools and each other. Ars Technica reports the flaw exposes how trust between agents and connected services can be exploited. Cybersecurity researchers are debating whether MCP was designed too loosely for sensitive, agent-to-agent workflows.

  4. 4
    Google's bug bounty flooded with AI-generated fake vulnerability reports●Google paid bug hunters to find open-source flaws, then AI flooded the queue with made-up bugsβœ‰newsTechnologySoftware39 min ago

    Google's bug bounty programme, which pays researchers to find security flaws in open-source software, has been inundated with fabricated bug reports generated by artificial intelligence. The wave of bogus submissions is overwhelming reviewers and complicating efforts to identify genuine vulnerabilities. Commentators say the case shows how AI tools are being used to game paid reporting schemes, straining security teams and threatening the usefulness of crowdsourced bug hunting.

  5. 5
    CrowdStrike: China-Based Suspect Used AI in South Korean Bank Hacks●CrowdStrike says China-based suspect used AI tools in South Korean bank hacksβœ‰newsTechnologyAI42 min ago

    Cybersecurity firm CrowdStrike says a suspect based in China used AI tools in hacks targeting South Korean banks. The finding highlights how artificial intelligence is increasingly being deployed in cyberattacks against financial institutions, and it is likely to intensify scrutiny of China-linked cyber activity and the security of South Korea's banking sector.

  6. 6
    AWS launches guide to building AI vulnerability harnesses●Building your AI vulnerability harness, Part 1βœ‰newsTechnologyCybersecurity41 min ago

    Amazon Web Services has published the first part of a technical guide on building an AI vulnerability harness, a framework for systematically testing AI systems for security flaws. The series walks practitioners through constructing tooling to probe AI models for exploitable weaknesses, reflecting growing industry focus on securing machine learning deployments against emerging attack vectors.

  7. 7
    US nearly provoked war with China over chatbot's fabricated claim, critics say●Another example of gross incompetence from techbros: "The U.S. almost started World War 3 with China by relying on.. fabMmastodonWorldUS Politics61 h ago

    Posts are circulating the claim that the United States came close to a serious confrontation with China after acting on fabricated information from a chatbot. The chatbot allegedly generated false reports that a Chinese vessel was carrying nuclear weapons components. Commenters are framing the episode as an example of gross incompetence in relying on AI tools for intelligence and national security decisions.

Repos