AI Briefing
GeekNews news
The latest AI and developer news from GeekNews, with short summaries.
Visit website →CLI Tool jevgrep for Coding Agents Enables Relevant Code Search via Natural Language Queries
A tool that returns relevant files and source code based on behavior-oriented queries in unfamiliar codebases.
buttonmasher: AI Agent Skill Simulating Impatient Users to Find Duplicate Action and Race Condition Bugs
The open-source skill simulates impatient user behaviors like double-clicking and retrying to find duplicate charges and state errors, supporting Claude Code, Codex, and Copilot CLI.
PickTalorys: Open-Source Personal AI Agent for Cloudflare Free Tier
Talorys deploys a private, single-user AI assistant entirely within a user's Cloudflare account using Workers, Durable Objects, and Workers AI, requiring no external servers or databases.
Porch: Open-Source Mac Menu Bar App for Summarizing Claude Code and Codex Sessions
The AGPL-3.0 licensed tool tracks token usage and generates daily, weekly, and monthly work summaries by analyzing local session records from Claude Code and Codex.
pg-jev: PostgreSQL Extension for Natural Language Data Search and Classification via TypeSafe Jev API
The extension enables SQL queries to evaluate natural language conditions against table rows using the TypeSafe Jev API, with configurable batching and session-based caching.
Tencent ARC Releases WorldCrafter: Open-Source Video World Model with Implicit 3D Memory
Tencent ARC Releases WorldCrafter: Open-Source Video World Model with Implicit 3D Memory
Lemma Open-Sources Platform for Shared AI Agent Workspaces Using Existing Subscriptions
Lemma has open-sourced a platform that lets teams use their existing AI subscriptions (like Claude Code or Codex) within a shared, permission-controlled workspace with multi-surface access.
New Paper Argues Lean Verification Fails to Guarantee Correctness of AI Autoformalised Navier-Stokes Proof
Researchers argue that semantic ambiguities in natural language make faithful AI autoformalisation computationally harder than the Halting problem, undermining confidence in the verification of OpenAI's claimed Navier-Stokes blow-up proof.
Author Finds Claude Opus 5.5 'Mesmerizing' for One-Shot 'Invisible Cities' Visualization, Though It Used More Time and Cost Than GPT-6 Astra
In a one-shot test, Claude Opus 5.5 generated a 'mesmerizing' 3D visualization of 'Invisible Cities' in 1 hour 25 minutes for ~$74, which the author preferred over GPT-6 Astra's 53-minute, ~$10 result, despite Opus using more time and resources.
Anthropic Engineer Clarifies Claude Code Prompt Suggestions Are Not Used for RLHF Training
An Anthropic engineer confirmed that prompt suggestions in Claude Code are designed for user flow, not to collect preference data for reinforcement learning.
PickTencentCloud Releases Octop, an Open-Source Self-Hosted Multi-Agent AI Assistant
The MIT-licensed platform supports multi-user isolation, local-first data storage, and integration with external coding agents like Claude Code and Codex via the Agent Client Protocol.
OpenRig: Open-Source Multi-Agent Orchestrator for Claude Code and Codex
OpenRig allows developers to define and boot multi-agent coding teams using YAML specs, managing sessions via tmux and a local daemon.
Agent Guard: Local Security Tool Prevents Secret Leaks in Claude Code and Codex
The open-source tool blocks risky file reads and scans Git commits locally using gitleaks to prevent API keys and tokens from entering AI agent transcripts or repository history.
a16z Report: Consumer AI Spending Concentrates in Top 1%, ChatGPT Leads Market
The report reveals that the top 1% of consumer AI spenders account for 19.5% of total observed spending, with an average monthly expenditure of $903.
OpenMuse: Open-Source Personal AI Agent with Browser and Terminal Capabilities
OpenMuse is an open-source personal AI agent built on CopilotKit that features a persistent browser, optional Linux terminal, and file management, with web search enabled by default via Parallel.
Offrun Offers Free Mac Workspace for Managing Multiple Coding Agents
Offrun is a free macOS application that manages existing coding CLIs like Claude Code and Codex in isolated git worktrees, featuring automatic account switching when usage limits are reached.
Edward Coristine Exposed to Potential Criminal Charges in At Least Six States Over Social Security Data
Whistleblower disclosures and DOJ admissions link Edward Coristine to the unauthorized copying of the NUMIDENT database covering over 300 million Americans, exposing him to potential criminal charges in at least six states.
PickOpenAI Safety Leader David Robinson Quits, Citing Broken Culture and Insufficient Care
David Robinson, who led safety reporting for ChatGPT releases, resigned via an essay in The Atlantic, arguing that AI firms are not being careful enough and calling for a cultural overhaul.
Tencent Zhuque Lab Releases AI-Infra-Guard v4.6.2 with Expanded CVE Coverage and New Agent Safety Benchmarks
The latest update adds 155 new CVE rules for 40+ AI components and introduces FORGE-Bench and RogueHandoff-20 to evaluate agent loss-of-control risks.
Mixdog: Open-Source AI Coding Harness Reduces Costs and Context Usage vs Native CLIs
Benchmarks on Terminal-Bench 2.1 show Mixdog achieves comparable accuracy to Codex CLI and Claude Code while reducing costs by 19-39% and context usage by 28-46%.