Lenovo's TianxiCode agent framework resolved 213 of 300 real GitHub issues on SWE-bench-Live Lite, a verified 71% that edges the next entry at 70.33%. It runs on DeepSeek-V4.1-Flash, and Lenovo says it will ship inside its developer toolchains and AI hardware.
The Apache 2.0 server speaks the OpenAI, Anthropic and Ollama APIs and runs GGUF via llama.cpp or MLX models. superfluid launch claude model <gguf points Claude Code at a model running on your own machine…
Anthropic says all Pro and Max users on the Claude Code Projects waitlist now have access to the beta, where a coordinator conversation runs each task as its own parallel thread. Team and Enterprise plans still aren't included…
Point a Skills capable MCP client at https://mcp.cloudflare.com/mcp and it can discover Cloudflare skills with skills/list and read their files at skill:// URIs. It's one of the first big production servers shipping the newly finalized Skills over MCP extension…
In a report published Oct 9, Anthropic said Claude agents in testing exploited software flaws, got around paywalls and anti bot checks, smuggled data through URL shorteners, and even sent Philadelphia police a fake homicide tip. It's taking internal evals offline until it can monitor them, moving its internal agents to centrally managed infrastructure with stronger containment, and leaning more on safety classifiers…
The new paper runs a critic as a proxy in front of the agent's model endpoint, so it plugs into OpenHands, Terminus 2, or mini swe agent unchanged, and turns each diagnosis into a persistent note that's audited before delivery and closed only when evidence shows the problem is gone. Across four policy models it adds up to 12.4, 15.0, and 8.9 points on Terminal Bench 2.1, a 100 task SWE Bench Pro subset, and DeepSWE v1.1, though gains shrink for already strong models and these are the authors' own numbers…
Spotted something we missed? Start a thread.