Tag
The author reflects on the decline of plan modes in AI-assisted coding, arguing that as models improve, these modes are becoming obsolete, while the need for human comprehension of software systems becomes more critical.
The author reviews five AI coding tools based on real-world production incidents and proposes a filter for safely integrating them into development workflows.
Promotion of the book 'GitHub Copilot and AI Coding Tools in Practice' by Nick Wienholt, which covers practical use of generative AI coding tools for software development across the full lifecycle.
A new Electron app consolidates subscriptions for Claude Code, Codex, and Cursor into a single agent-agnostic interface, allowing any model to use agent spawn and maximize subscription usage.
A detailed exploration of background coding agents—AI agents that work autonomously in cloud sandboxes and open pull requests—contrasting them with autocomplete and IDE-based tools, based on FactoryKit's real-world experience shipping over 180 features in two weeks.
The author compares numerous AI coding harnesses, sharing personal experience with Hermes, Cline, Cursor, and others, noting that Hermes paired with Aphrodite works well for medium-to-large projects.
A comparison article pitting AI coding tools Claude Code, OpenCode, and Pi against DeepSeek V4 Flash in a harness showdown.
A user expresses interest in building a software factory, experimenting with AI coding tools like Cursor, Amp, Droid, and Devin, citing a podcast where Greg Isenberg discusses improving workflow beyond a single Claude Code instance.
The author built and shared a personal reference document comparing free tiers of 140 AI coding tools, aiming for honest, up-to-date information without sponsored placements.
A study comparing Claude Code and OpenCode reveals that Claude Code sends 33k tokens before reading the prompt while OpenCode sends only 7k, highlighting significant inefficiency in Claude Code's cache strategy and token usage.
Analyzes the competition between Cursor and Claude in the coding field in early 2025. Cursor was surpassed after Claude trained on its data. Musk then supported the Cursor team and released the Grok 4.5 model through SpaceXAI, changing the competitive landscape.
Researchers have devised a pull-based prompt injection attack called HalluSquatting that exploits AI coding assistants' tendency to hallucinate resource identifiers, enabling the assembly of massive botnets and large-scale attacks.
SigMap is an open-source grounding layer for AI coding agents that provides a deterministic map of code repositories, reducing wasted context and improving retrieval accuracy and task success rates.
The article argues that codegraphs address the wrong aspects for improving AI coding agents, suggesting a misdirection in current development efforts.
The Every team publicly shares its compound engineering methodology and open-sources supporting plugins. Through a four-step cycle and multiple AI agents working in parallel, a single-person team manages 5 products, spending 80% of the time not writing code.
Cursor has launched a mobile app for iOS that allows users to prompt and interact with its coding agents from their phones, following a trend of AI coding tools moving to mobile.
User shares experience switching back from Codex to Claude Code, noting significant differences between the two.
A junior developer reflects on whether it is normal to find understanding and reviewing code changes more challenging than writing code when using AI coding tools.
The article discusses why enterprises adopt AI coding tools like AWS Kiro, GitHub Copilot, and Cursor even when they rely on Claude as the underlying model, focusing on enterprise needs such as security, compliance, and workflow integration.
An analysis of how developers choose between AI coding tools in 2026 as their feature sets become similar.