New research from the United States concludes that speaking to AI models such as ChatGPT or Gemini produces worse results than typing prompts, even when typed prompts contain mistakes. The study found…
A new benchmark study from a40-labs measures three shapes of agent memory—file-based curation, structured stores, and trained experience—and finds that the file-based approach used by Claude Code, Cli…
Progressive context shaping, a method of continually updating a small current-state file rather than relying on a static prompt, helps AI coding agents stay on track during multi-hour or multi-day run…
Anthropic rolled out invisible watermarking in Claude-generated text and file outputs from August 2, 2026, to comply with the EU AI Act's Article 50(2) Code of Practice on Transparency of AI-Generated…
Spotify launched Xirp on August 10th, an agentic development environment that orchestrates AI coding agents across models, repositories, and parallel sessions without model lock-in. Spotify's head of …
Developer Arpit Uppal has published a practical guide to AGENTS.md, a markdown file that AI coding agents read at session start to understand a repository's commands, conventions, and architecture. Up…
Anthropic announced Tuesday that it will embed invisible watermarks and digitally signed metadata into text and images generated by its Claude AI models, including Claude Code and Claude Cowork, to co…
Anthropic's Claude Code, introduced on May 22, 2025, has fueled a wave of low-quality 'vibe coded' game projects, including an AI-translated English patch for the Dreamcast RPG SEGAGAGA that drew crit…
A developer has released agent_drift, an open-source Python script that detects when copies of AI agent instruction files such as CLAUDE.md, .cursor/rules/*.mdc, and .github/copilot-instructions.md ha…
A working prototype plugin for Claude Code, named Claudish to English, rewrites assistant messages into plain English using a local LLM via ollama, and is display-only, preserving the original text in…
Claude Code, an AI coding agent from Anthropic, solved a complex math problem that had stumped a developer for an extended period by identifying a symmetry in variables that simplified a multi-step it…
A developer explains the current state of coding agent context files, noting that AGENTS.md has become an open, vendor-neutral standard backed by OpenAI, Google, and the Linux Foundation's Agentic AI …
Anthropic announced that text and files generated by its Claude AI models launched on or after Aug. 2 will include watermarks to comply with EU regulations on AI transparency. The watermarking will ap…
A new open-source Agent Skill called log-decisions provides an append-only DECISIONS.md journal for AI agents to record consequential judgment calls, with decide/assume/escalate rules. The skill, comp…
Anthropic's Claude Code, an AI coding tool, may contain security backdoors that could compromise system integrity during deployment, according to a security analysis. The tool's ability to execute she…
A developer has created a tool to make project repositories self-sufficient by including source code, goals, issues, and documentation, with seamless integration for AI agents like Claude Code. The to…
Developers have released open-source cost-tracking tools including Databricks Cost Optimizer and Ungate in response to reports of high token consumption in Claude Code, with users reporting spending $…
A developer has released Overcap, an open-source tool that tracks AI usage across coding tools like Claude Code, Codex, and Grok, available for Windows, Linux, and Mac. It addresses the gap left by th…
Firecrawl released anydoc, a Rust library that converts Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, and PDF files into GitHub-Flavored Markdown in single-digit milliseconds, with bindings f…
Anthropic will begin watermarking AI-generated content from Claude, Claude Code, and its API, as well as Claude models via third-party services, to comply with the EU's AI Regulation, with global impl…