Tag
Claude Brain is an open-source tool that adds persistent memory to Claude Code with a single file, automatically recording decisions, bugs, and plans, with local search support and no need for a database or cloud service.
Impeccable 4.1 releases with 43 PRs, improving critique reliability, design-system consistency, native app review, plugin installs, and live mode support across more dev environments.
bismar is a new terminal-based source code browser that lets you open, browse, diff, and bundle packages from npm, jsr, crates.io, PyPI, gems, Packagist, GitHub, and more, with a focus on security and machine-friendly output.
The author shares 'diff-rationale', a CLI that records the reasoning behind file changes via git, allowing AI agents like Claude to preserve rationale across sessions and query it later.
sqlite-utils 4.2.1 fixes a crashing bug caused by an unlisted typing-extensions dependency, and adds a smoke test to verify CLI works without dev dependencies.
DeepSeek Harness has been released, featuring a trace mode that visualizes Agent's thinking steps, tool calls, and return results, and uses colors to distinguish different types of content. It can be installed via npm and is suitable for developers to observe Agent work progress.
Bullet is a faster coding agent for macOS and Linux that optimizes AI tool calls by routing tasks to appropriate models, using targeted file reads, and executing operations in parallel. It's launching in private beta, free to use, after being built out of frustration with slow agent runs.
A tweet highlights that Claude Code sessions can now be named and message each other directly, allowing developers to coordinate multiple coding sessions via CLI commands.
A technical primer on using GitHub's stacked pull request feature with the Jujutsu version control system, covering prerequisites, jj configuration, and manual workflows for creating and linking stacks.
Simon Willison announces the alpha release of alchemy-utils, a database-agnostic Python library and CLI inspired by sqlite-utils, built with AI assistance and supporting PostgreSQL, SQLite, and DuckDB.
Cursor's engineering blog describes how they built a development environment for cloud agents, introducing the anydev CLI and the Cloud Doctor self-healing mechanism, enabling agents to now contribute more than half of merged PRs.
Claude Code 2.1.228 released with 18 CLI changes, including hardened claude.ai-synced skills that no longer shadow local commands, and grep now using ripgrep for faster project-wide searches.
Browser Use promotes the Hermes agent with cloud browsers for stealth browsing, offering $15 free credit at $0.02/h pricing, and highlights Browser-Use CLI 3.0's backend delivering 60% less token spend.
mcptoon is a zero-dependency Python CLI client for MCP servers that uses a compact TOON format to reduce tool discovery tokens by up to 97%, saving context window space for AI agents.
Vercel CEO rauchg praises deepsec as an essential security review tool, noting it has become a verb within Vercel, and highlights the new 'npx deepsec init' command for one-command security reviews.
Ante is a self-contained coding agent in a single ~15MB Rust binary from Antigma Labs, designed to run entirely offline with built-in inference. It boasts strong Terminal-Bench 2.1 results, low resource usage compared to Claude Code, and supports GGUF models.
NotesQR, an anonymous P2P file sharing tool built on WebRTC, has been completely rewritten with faster handshakes, plus a new CLI and MCP integration so AI agents like Cursor or local scripts can send and receive files programmatically.
Soup is a new CLI tool that simplifies LLM fine-tuning to one command, with templates for chat, code, tool-calling, and more, plus layer streaming to run on modest GPUs.
Slim is a minimalist command-line tool that converts local ports into clean HTTPS domains (e.g., myapp.test) with one command. It also supports intranet penetration to generate public links for collaborative debugging and demos. It's completely free and open source.
Kane CLI lets developers run natural language browser and mobile app tests directly from the terminal.