AgentGauntlet, a new testing tool for AI agents, simulates real-world chaos like context drops, tool timeouts, and bad API data to evaluate agent resilience, reporting a 0% resilience score when an ag…
A developer's evaluation of three agent memory frameworks—file-based, structured store, and reinforcement-learning-trained experience—found that the file-based approach, as implemented in OpenClaw's m…
Pakistan's Higher Education Commission (HEC) will require all university students to complete a mandatory three-credit-hour AI course by Fall 2026, but its draft generative AI policy banning AI-genera…
Mnemon-dev released mnemon 0.1.0, an LLM-supervised persistent memory system for AI agents that provides graph-based recall and cross-session knowledge in a single binary, compatible with DeepSeek Har…
Taiwan's Ministry of Digital Affairs (MODA) confirmed that overseas hackers used AI agents to target government agencies in a cyberattack last month, compromising at least 85 government user accounts …
A solo developer has built a Claude Code plugin that queries 10.6 million earnings-call embeddings, enabling users to ask natural-language questions about earnings transcripts and receive scheduled re…
A developer has open-sourced OpenClaw Control Plane, a TypeScript monorepo that adds a governance layer to OpenClaw instances on Railway, addressing the problem of 'agent sprawl' in workplace AI. The …
Prof. Tom Yeh's AI by Hand newsletter published a series of seminars and tutorials on AI topics, including a July 8, 2026 seminar on Qwen 3.6, a May 7, 2026 seminar on Gemma 4, and a March 2, 2026 sem…
In early July, attackers used open source AI agents to autonomously hack government systems and energy companies, signaling that AI-powered attacks against critical infrastructure are no longer theore…
Taiwan's Ministry of Digital Affairs reported that an AI-assisted cyberattack targeted government agencies in July, with affected bodies completing incident handling, while reports from The Register, …
A developer proposes that MCP servers need a 'capability budget' in addition to authentication, defining a short-lived contract that constrains tool actions, targets, resources, quantity, and expiry. …
Cybersecurity firm Dream reported that AI agents built on open-source Hermes and OpenClaw frameworks breached Taiwanese government systems over four days in early July, compromising 85 government user…
SKALE Labs launched AgentPit, a prediction market sandbox on its zero-gas blockchain that replicates Polymarket's mechanics, giving developers a risk-free environment to train AI trading agents with $…
A study posted on arXiv in late July found that AI systems are not yet capable of fully automating open-ended research, with an AI system built by Princeton University researchers scoring 2/6 and 1/6 …
GitHub's Secure Open Source Fund invested more than $500,000 across 50 projects in Session 4, pairing maintainers with GitHub Security Lab experts, security tools, and AI-assisted workflows. The progr…
Echo, in partnership with NanoClaw, eliminated 1,400 CVEs from NanoClaw's container images by using vulnerability scanners, upgrading safe libraries, backporting patches, and replacing the base OS wit…
Autonomous AI agents built on open-source frameworks breached Taiwanese government systems, compromised credentials, and probed a nuclear safety agency in a multi-day cyberattack, according to researc…
DeepSeek is recruiting researchers, engineers, and product staff for an Agent Harness team in Beijing and Hangzhou, according to its hiring page and reports from May 20–21, 2026. The roles focus on co…
Suspected Chinese cyber operatives used publicly available AI tools to compromise Taiwanese government systems, including its nuclear safety agency, supply-chain vendors, and at least seven energy com…
Researchers at Israeli cybersecurity firm Dream said suspected China-linked hackers used an open-source AI-agent system to conduct what may be the first observed end-to-end autonomous cyberattack agai…