Tag
Augment's Cosmos platform enables AI agents to handle the full software development lifecycle, dramatically reducing code review time from hours to minutes by automating routine checks and focusing humans on design decisions.
Shared a practical website that collects ready-made Agent Loop templates. The core idea is to let the AI agent automatically complete tasks. It includes common loops like Test Until Green, and is suitable for use with Claude Code and Cursor.
A guide on building autonomous engineering pipelines, covering integration with services like Slack and GitHub, and highlighting Devin's built-in capabilities for rapid setup.
Claude Managed Agents can now operate in a user-controlled sandbox on your own infrastructure, with new integration guides for Blaxel AI, e2b, Google Cloud, Namespace Labs, and Superserve AI.
A detailed comparison between Nix Flakes and their equivalents in the Guix package management system, covering dependency declarations, pinning, purity, outputs, development environments, and system configuration.
DevCleaner is a tool that helps reclaim gigabytes of storage space by cleaning up files from development tools and AI applications.
Boris Cherny and Cat Wu of Claude Code recount the first year of the product, highlighting the shift from humans talking to agents to humans talking to loops that drive agents, and how routine agents now handle code review and bug fixes autonomously.
npm v12 introduces security-related breaking changes to npm install that disable automatic execution of scripts, git dependencies, and remote URL dependencies by default. Users can prepare by upgrading to npm 11.16.0+ and reviewing warnings to explicitly opt into trusted behaviors.
ACP provides a unified interface to access multiple AI agents including Codex, Claude, OpenCode, Devin, and Gemini, along with ~40 others, from a single app.
Describes a loop command in Cursor to automatically fix flaky tests by running the test suite multiple times, collecting intermittent failures, and fixing or quarantining them until five consecutive green runs.
A blog post from Posit defends YAML against the prevailing consensus that TOML is superior, tracing the history of configuration formats and arguing that YAML's specification and tooling have evolved to address past criticisms.
Lovable has shipped a new security scanner that runs before every deploy, catching misconfigurations, missing RLS policies, and cloud gaps, with automatic fixes and deep scan capabilities.
A technical exploration showing how Nix can build a Guix derivation, highlighting the shared underlying 'Input Output Machine' architecture and the possibility of cross-ecosystem interoperability.
Announcing that the Compound Engineering plugin repository has reached 20,000 stars on GitHub. It provides AI agents and skills to make each unit of engineering work easier, focusing on planning and review to avoid technical debt.
Hugging Face's hf CLI is shown to be far more token-efficient and successful for AI agents than hand-rolling raw API calls, with benchmarks showing up to 6x fewer tokens and 94% vs 84% task success, demonstrating that good abstractions are cached intelligence for agents.
Boxes.dev allows you to run Claude Code and Codex in your own cloud environment.
A discussion about preserving the debugging path between coding agent runs, referencing the Visr tool that keeps trajectory alongside diffs, and asking how others handle this.
oh-my-grok is a lightweight productivity plugin for the Grok Build CLI, adding skill gates, Ralph/Ultrawork loops, todo continuation, and handoff summaries. It is an open-source, early-stage tool that fills the plugin gap.
Cognition introduces Devin Desktop, a unified interface for managing fleets of local and cloud AI agents, with support for any ACP-compatible agent and an integrated IDE.
SLAX is an alternate syntax for XSLT that uses a C/Perl-like syntax to improve readability and maintainability. The libslax open-source implementation parses and executes SLAX files.