Anthropic shipped a third permission policy called `auto` for Claude Managed Agents on September 10, routing each tool call to a server-side evaluator that returns allow, ask, or deny. Anthropic said …
Docket, a new tool from developer Dillonsmart, captures coding-agent session data and folds it against the commit diff to produce a per-hunk evidence record, flagging lines no human has reviewed. The …
Anthropic's Claude agent can be built into a self-correcting AI agent using a Python script that generates code, runs it against hidden tests, and retries with a critic's feedback until tests pass or …
Two major AI writing leaderboards, EQ-Bench Creative Writing v3 and arena.ai's Creative Writing category, produce inverted rankings for the same models, with claude-fable-5 ranked #1 on arena.ai but 6…
K-Bench 01, an evaluation built from first-turn requests sampled from live user traffic on K-Dense Web, ran 1,602 completed agent runs by nine frontier models in identical sandboxes, with three blinde…
A developer's tests show that Claude Code's internal configuration layers fail to prevent the coding agent from writing files, while a macOS sandbox profile consistently blocks it. Across measurements…
Nexla's task-specific MCP server outperformed Avoma's MCP server in a benchmark on the same Avoma workspace, correctly answering all call analysis questions while Avoma scored 0.00 on three measured t…
Priya Nair published a tutorial for building a Python token usage dashboard that logs tokens and dollar cost for every OpenAI, Anthropic, and Gemini API call into SQLite and visualizes spend via a Str…
Simon Willison released llm-anthropic 0.26, a plugin for the LLM command-line tool, adding support for Claude 5 models (claude-fable-5, claude-sonnet-5, claude-opus-5) and server-side tools including …
Anthropic's Claude API users can cut their bills by up to 90% by combining prompt caching and the Message Batches API, according to a tutorial by Rachel Goldstein. Prompt caching bills repeated contex…