AI Briefing#openclaw
The latest AI and developer news about #openclaw, with the original source and a short summary.
The latest AI and developer news about #openclaw, with the original source and a short summary.
QClaw-4B has been released as a 4B model for OpenClaw-compatible agents.
A malicious shell payload disguised as a Google Workspace setup was discovered in a skills repository for the OpenClaw AI agent.
Over 40,000 OpenClaw instances have been exposed externally, putting them at risk of RCE attacks.
The OpenClaw tool for Anthropic's Claude models is supported again, supporting the latest features such as Claude 4.6's adaptive thinking and prompt caching.
Three security vulnerabilities were discovered in OpenClaw, including a sandbox bypass with a severity of 9.9.
An analysis of the Claude Code source code lays out the core design principles and open challenges of AI agents.
Behind the glamorous success story, security incidents and scalability problems have come to light.
Introducing Nilbox, which passes fake tokens and substitutes them with real tokens at the network layer.
A log-compress-atomize-link structure improved OpenClaw's memory reusability.
Among 138 CVEs in the OpenClaw ecosystem, the entries critical for self-hosting were singled out as warnings.
If the gateway won't come up after the 4.14 update, suspect a tangled entrypoint path.
By splitting work between GLM and Sonnet, OpenClaw operating costs were cut to under $30 per month.
Real users shared concrete details about OpenClaw's use cases, costs, and reliability.
Ran Gemma 4:26B on a Mac Studio M4 Max using OC + LM Studio, achieving 40 tok/s.
Fixing the cache settings dropped OpenClaw's per-call cost from 13 cents to 2 cents.
A 2-minute morning check routine for an actually operating openclaw agent.
After comparing several LLMs, Gemma 4 was found to be the best for agentic workflows.
openclaw 4.14 conflicts with lossless-claw, causing the agent to freeze.
Shares how to split up models by role for OpenClaw based on PinchBench scores.
Shares recommended settings and reasons for using GPT 5.4 in OpenClaw.
GitHub has released Secure Code Game Season 4, which teaches the vulnerabilities of agentic AI.
Microsoft is testing an OpenClaw-style Copilot agent with stronger enterprise-grade security controls.
In an OpenClaw+Ollama setup, glm-4.7 was more stable for agentic tasks than Qwen or Gemma.
The overnight memory feature worked well, but it exposed weaknesses in memory protection, verification, and write methods.