Open-source projects are formalizing rejection of AI-generated contributions at an accelerating pace, with five independent actions in seven days. Codeberg banned AI-generated code in a 358-144 member…
Codeberg voted 358 to 144 (71% in favor) to ban projects "mostly" written by generative AI tools, with the policy now live in its Terms of Use. The ban prohibits repos where AI wrote the majority of c…
Infrastructure developer Earendil Works gave its Pi coding agent a single tool called fabric_exec, which accepts TypeScript programs that call Pi's capabilities as an API, and one such program ran con…
A year after drawing a clear line between "agentic engineering" and "vibe coding," Simon Willison now acknowledges that coding agents have become so reliable that he no longer reviews every line of co…
Pi, a coding agent built by Mario Zechner and acquired by Earendil Inc., deliberately omits features like MCP support, sub-agents, and plan mode that competitors emphasize, instead offering only four …
A developer attending Local-First Conf in Berlin describes a widespread dissonance among engineers who use large language models despite agreeing with critics' concerns about slop, copyright, environm…
Databricks CTO Matei Zaharia's team built an internal benchmark on their own polyglot codebase to evaluate coding agents, finding that multiple models are now competitive at the top tier, including op…
A new analysis compares large language models to sea squirts, arguing that post-training optimization for specific tool shapes is creating a Darwinian niche where models shape their own environment. T…
A software engineer argues that AI coding agents produce overly defensive, complex, and underspecified code because they are trained to never refuse a request, leading to hidden decisions and endless …
Newer Anthropic Claude models, including Opus 4.8 and Sonnet 5, are generating malformed tool calls for Pi's custom edit tool by inventing extra fields, a problem absent in older models. Armin Ronache…
GitHub faces stability issues and high-profile departures like Ghostty, prompting developers to seek alternatives. Tangled, a Git forge built on the AT Protocol, offers a decentralized, open-source so…
Three friends in Vienna used Anthropic's Claude AI to create VibeTunnel in June 2025, sparking a chain of agentic coding projects that culminated in Peter Steinberger's OpenClaw, which exploded in pop…
Multi-agent AI systems are achieving significant throughput gains by running multiple autonomous coding agents in parallel, with teams like Geoffrey Huntley's Ralph Loop and Steve Yegge's Gas Town dep…
Seven senior software engineers at Kilo who build and use AI coding agents report that running 2-4 "foreground" agents in parallel is the productive sweet spot, contradicting claims of managing 50-100…
Software engineers Mario Zechner and Armin Ronacher, who helped build the OpenClaw AI agent, warned that AI coding tools are flooding software with "vibe slop" by enabling users to skip critical devel…
Software developer Armin Ronacher criticized the growing trend of AI-generated "slop" bug reports on open-source projects, saying users are submitting issues that lack their own voice and contain inac…
The article describes a small Rust-based back-end system called "upvoters," created to handle anonymous upvotes on static blogs, with compatibility for Hugo. The author also mentions recent website de…