MindMux presents brain.md, an open-source toolkit that provides a persistent memory layer for coding agents by storing project knowledge as plain Markdown files within a repository. This system ensures that critical decisions and constraints are durable across different LLM sessions and machines via version control. A zero-dependency CLI manages the reading and writing of these files to maintain data integrity through an append-only timeline.
- Uses Markdown instead of databases like SQLite to facilitate easier diffing in git history.
- Features a "correct by construction" design that prevents malformed edits by making the CLI the exclusive writer.
- Supports integration with several agents including Claude Code, Codex, Cursor, and Pi.
- Uses Markdown instead of databases like SQLite to facilitate easier diffing in git history.
- Features a "correct by construction" design that prevents malformed edits by making the CLI the exclusive writer.
- Supports integration with several agents including Claude Code, Codex, Cursor, and Pi.
The author explains how enabling the built-in sandbox feature in Claude Code has transformed their productivity by allowing for an autonomous workflow. By activating auto-allow mode via the /sandbox command, users can permit AI agents to execute repetitive tasks like running tests or installing dependencies without needing constant manual approval for every individual command, which reduces mental overhead and time spent waiting on prompts.
- Claude Code provides a built-in sandbox that uses macOS Seatbelt on Mac and Bubblewrap on Linux/WSL2.
- The environment restricts file writing primarily to the current project directory and temporary session files.
- Network access is controlled, requiring manual approval when the agent attempts to connect to a new domain.
- For maximum security and complete isolation from a system, using Docker containers or virtual machines remains more robust than the built-in sandbox features.
- Claude Code provides a built-in sandbox that uses macOS Seatbelt on Mac and Bubblewrap on Linux/WSL2.
- The environment restricts file writing primarily to the current project directory and temporary session files.
- Network access is controlled, requiring manual approval when the agent attempts to connect to a new domain.
- For maximum security and complete isolation from a system, using Docker containers or virtual machines remains more robust than the built-in sandbox features.
This article explores the concept of harness engineering, arguing that a functional AI agent is defined not just by its underlying model, but by the scaffolding built around it—including prompts, tools, sandboxes, and feedback loops. The author suggests shifting focus from picking the smartest model to designing robust systems that turn raw models into reliable agents. By treating mistakes as signals for new constraints rather than simple failures, engineers can create a ratchet effect that continuously improves agent performance through better configuration.
Main topics:
- Defining an agent as the combination of a model and its harness
- Reframing model errors as configuration or skill issues
- Using failure history to implement permanent rules via hooks and documentation
- Core primitives including filesystems, bash execution, sandboxes, and memory management
- Managing context rot through compaction and tool offloading
- Achieving long-horizon work through planning, verification, and agent splits
Main topics:
- Defining an agent as the combination of a model and its harness
- Reframing model errors as configuration or skill issues
- Using failure history to implement permanent rules via hooks and documentation
- Core primitives including filesystems, bash execution, sandboxes, and memory management
- Managing context rot through compaction and tool offloading
- Achieving long-horizon work through planning, verification, and agent splits
Simon Willison details creating a custom macOS presentation app, "Present," in just 45 minutes using Swift and SwiftUI. Frustrated with the risk of browser crashes when presenting a series of web pages, he built an app that displays URLs as slides, offering features like full-screen mode, keyboard navigation, and automatic URL saving. He even added remote control functionality via a web server and Tailscale.
The project highlights the power of AI-assisted coding and expands his skillset, demonstrating how experienced software engineers can quickly learn new languages and tools to solve personal problems. The resulting app is a simple, effective solution tailored to his specific needs.
The project highlights the power of AI-assisted coding and expands his skillset, demonstrating how experienced software engineers can quickly learn new languages and tools to solve personal problems. The resulting app is a simple, effective solution tailored to his specific needs.
Goose is a free, open‑source AI agent that runs locally and can autonomously plan, code, test, debug, and execute full development workflows—making it especially useful for data scientists who need to automate repetitive, multi‑step tasks. It supports any LLM, interfaces with file systems and APIs, and can extend its capabilities via the Model Context Protocol (MCP) to connect with databases, Git, Slack, and more.
- Autonomous task execution from high‑level instructions.
- Local execution preserves data privacy and control.
- LLM‑agnostic: works with GPT‑4, Claude, or local models.
- Two interfaces: desktop GUI and CLI.
- Extensible through MCP for external tools and services.
- Ideal for rapid prototyping, data pipeline automation, MLOps, and environment setup.
- Autonomous task execution from high‑level instructions.
- Local execution preserves data privacy and control.
- LLM‑agnostic: works with GPT‑4, Claude, or local models.
- Two interfaces: desktop GUI and CLI.
- Extensible through MCP for external tools and services.
- Ideal for rapid prototyping, data pipeline automation, MLOps, and environment setup.
A new ETH Zurich study challenges the common practice of using `AGENTS.md` files with AI coding agents. LLM-generated context files decrease performance (3% lower success rate, +20% steps/costs).Human-written files offer small gains (4% success rate) but also increase costs. Researchers recommend omitting context files unless manually written with non-inferable details (tooling, build commands).They tested this using a new dataset, AGENTbench, with four agents.
A guide to supercharging Claude Code with Skills and the Model Context Protocol (MCP), including running Claude Code in an IDE like Cursor or VS Code. It covers setting up Skills, connecting to MCP servers, and combining both for powerful workflows.
Vercel's research shows that embedding a compressed 8KB docs index in AGENTS.md achieves a 100% pass rate for Next.js 16 API evaluations, while skills maxed out at 79%, even with explicit instructions. This suggests that passive context provision via AGENTS.md is more effective than active retrieval with skills for framework-specific knowledge in AI coding agents.
Zhipu AI has released GLM-4.7-Flash, a 30B-A3B MoE model designed for efficient local coding and agent applications. It offers strong coding and reasoning performance with a 128k token context length and supports English and Chinese.
A simple, open format for guiding coding agents, used by over 60k open-source projects. It's a dedicated, predictable place to provide the context and instructions to help AI coding agents work on your project.
