Full-text search across 58884 articles. Combine with topic and date filters; results sorted by relevance.
A developer built Codelux, a local daemon that gives AI coding agents persistent project memory across sessions via MCP on localhost, indexing a 120-file Go project in about three seconds and cutting tokens on a single f…
Anthropic's Claude Code 2.1.280, documented as of September 23, 2026, treats skills as folders containing a SKILL.md file whose name and description load into context every turn while full instructions load only on a mat…
A developer has built Tokenectomy, a preprocessing layer that filters logs, stack traces, and tool output before they reach an AI coding agent, aiming to stop agents from re-investigating already-fixed issues during long…
A developer argues that AI coding tools have largely solved the "how to build" problem while leaving the harder question of "what to build" unanswered. The engineer recounts spending a weekend using AI to ship a working …
A developer argues that as AI-assisted coding lowers the cost of shipping small product changes, testing becomes the critical feedback loop that keeps that speed safe. The piece contends that AI can accelerate implementa…
Dimitris Mourtzis released OtoDock 1.6.0 on September 9th, a self-hosted platform that runs persistent AI agents such as Claude Code and Codex across company departments, with features including broader local-model suppo…
Mohammed Nihal03 released AI Agent Observatory, an open-source local dashboard that analyzes Claude Code sessions to score agent health from 0 to 100 and flag degrading behavior, with the tool catching its own author edi…
Andrew Ng, founder of DeepLearning.AI, published a spec-first workflow for coding agents on September 4, urging developers to treat them as collaborators requiring repeated verification rather than unattended systems. Th…
A preliminary study by Pengcheng Xu, published by AgentConnect, found that coding agents often prefer grep over Language Server Protocol (LSP) semantic navigation for symbol localization tasks, with semantic navigation u…
An analysis of 6,876 Reddit posts and comments across 266 subreddits over six months found that developers are increasingly routing different AI coding models to different tasks, driven by workflow and cost concerns rath…
Sonoma County supervisors approved a settlement paying $50,000 to each of three residents and enacting regulatory reforms to end the ACLU lawsuit over drone surveillance used for code enforcement. The county admitted no …
An analysis of the Claude Code 2.1.257 binary reveals a new 'Fable 5.1 prompt bundle' capability assigned to Anthropic's Fable 5.1 and Mythos 5.1 models, which share the same underlying model. The bundle includes a syste…
A developer's attempt to run an AI coding agent entirely offline on a laptop using the open-source pi.dev agent CLI, Ollama, and Alibaba's Qwen3 27B model succeeded in building a 1,086-line Tetris game in plain HTML/CSS/…
Anthropic's Claude Code can consume far more tokens than expected because each request includes system prompts, project instructions, tool definitions, and conversation history, with prompt caching reducing costs only wh…
Security engineer David Seagal has released VibeGuard, a free open-source security linter designed specifically for AI-generated code. In a real production codebase audit, the tool scanned 533 files and found three criti…
Anthropic's Claude Code tool result caching reduces token consumption by 40-60% in multi-hour sessions by storing and reusing file reads and shell command outputs, according to an article written with AI assistance. The …
Veryfront released Veryfront Code, an open-source full-stack framework for building AI apps and agents with TypeScript and React, supporting Node.js, Deno, and Bun runtimes. The framework includes features such as agents…
Baro, a command-line software factory built on the TypeScript runtime Mozaik, runs concurrent coding agents that plan a goal into bounded units of work and execute it via an event bus with participants including a planni…
A developer proposes a 'context packet' design for coding agents, arguing that giving an agent access to an entire repository can reduce understanding due to low signal density. The packet is a small, versioned, machine-…
A developer built a circuit breaker for Claude Code's weekly quota after 15 LaunchAgent jobs silently failed for six hours, costing ¥1.2M/month in lost automation. The tool, claude-quota-guard.py, detects quota-limit err…