AGENTS .md
PulseAugur coverage of AGENTS .md — every cluster mentioning AGENTS .md across labs, papers, and developer communities, ranked by signal.
27 day(s) with sentiment data
-
Claude Code and Codex differ in monorepo agent file handling
This article compares how Claude Code and Codex handle agent configuration files within monorepos. Codex starts at the project root and walks downwards, prioritizing files closer to the working directory, and can be inf…
-
User switches from Claude Code to Pi citing stability issues
A user has decided to switch from Claude Code as their primary coding assistant to Pi, citing numerous small frustrations with Claude Code's stability and feature implementation. Despite Anthropic's recent efforts to im…
-
RepoXray tool audits code repos for AI slopsquatting and secret exposure
RepoXray is a new tool designed to audit code repositories before installation or use with AI coding agents. It addresses concerns such as AI slopsquatting, where attackers register packages that LLMs hallucinate, by ch…
-
AI coding assistants exploited by new AGENTS.MD virus targeting Markdown files
A new type of computer virus, dubbed AGENTS.MD, is exploiting AI coding assistants like Claude Code and Codex by disguising malicious instructions within seemingly harmless Markdown files. Attackers embed these files in…
-
Developer details effective AI instruction placement for Claude Code
A developer shares a strategy for effectively instructing AI models like Claude Code by placing rules in specific locations within configuration files. The developer emphasizes that rules placed deeper in the file hiera…
-
Keboola team boosts dev productivity with AI tools, not just models
A software development team at Keboola has significantly increased its productivity by building tools and processes around existing AI models like Claude and Gemini, rather than relying solely on the models themselves. …
-
Coding agents vulnerable to malicious instruction files in Downloads folder
Security researchers have identified a vulnerability where malicious Markdown files, such as AGENTS.md or CLAUDE.md, can be silently downloaded into a user's Downloads folder. When coding agents like Codex or Claude Cod…
-
Harness engineering grants AI agents autonomy safely
The article discusses "harness engineering" as a method to grant AI agents more autonomy without compromising safety. It argues that prompts alone are insufficient for reliable agent operation, and instead, a surroundin…
-
Best practices for AI coding agents: reduce tokens and bugs
To improve the reliability and efficiency of AI coding agents, it's recommended to assign one task per session and maintain clear separation between unrelated work. Developers should also limit documentation like AGENTS…
-
Aletheia AI coding tool releases patch with npm server start fix
Aletheia, a local memory and rules hub for AI coding tools, has released version 0.1.13. This patch includes the ability to start the MCP server directly from npm, addressing an issue where the MCP Registry entry was no…
-
Aletheia AI coding tool releases patch update with improved docs and registry publishing
Aletheia, a local memory and rules hub for AI coding tools, has released version 0.1.12. This patch release includes updates to its documentation, moving it to a dedicated website and providing individual release notes …
-
Developer integrates Claude Code, Codex, and Gemini CLI for unified repository work
A developer has outlined a method for using multiple AI coding agents, specifically Claude Code, Codex, and Gemini CLI, within a single repository. The approach involves centralizing instructions in a shared file (AGENT…
-
Integrating CLAUDE.md and AGENTS.md for Project Management
This article explores how to integrate CLAUDE.md and AGENTS.md within a single project. It suggests using shared project rules for Claude Code and Codex, incorporating a small import when necessary, and maintaining a se…
-
Malicious markdown files hijack AI coding agents like Codex and Claude
A security vulnerability has been discovered where malicious markdown files, such as AGENTS.md or CLAUDE.md, can be placed in an AI coding agent's working directory to hijack its operations. These files trick agents lik…
-
Consolidate coding agent rules for Cursor, Claude Code, and Codex
This article explains how to manage coding instructions across multiple AI agents like Cursor, Claude Code, and Codex. It details how each tool interprets rule files, such as AGENTS.md, CLAUDE.md, and .cursor/rules, and…
-
LLM skill files: Shell version yields similar gains to full structure
A recent retest of LLM skill files revealed that while Kimi K3.1 Pro and DeepSeek V4 Pro showed gains with a full skill file, a simplified "shell" version of the file yielded similar improvements. This suggests that for…
-
AI coding agents face security risks from exposed secrets and malicious instruction files
Two security vulnerabilities are highlighted for AI coding agents, focusing on how they handle sensitive information and execute instructions. The first issue, "MCP security for coding agents," details how secrets like …
-
Coding agents perform better with fewer, more accurate docs, author argues
The author argues that providing coding agents with more documentation is counterproductive, as outdated or inaccurate docs can actively harm their performance. Instead, they suggest that agents need fewer, more accurat…
-
Git worktrees enable parallel AI coding agent use in single repositories
This guide explains how to use Git worktrees to manage multiple AI coding agents working on the same repository simultaneously. By creating separate working directories for each agent, developers can prevent conflicts a…
-
New method enables AI coding agents to share team context
A new method for AI coding agents to share context across a team involves splitting information into two parts: instructions and growing knowledge. Instructions, such as build commands and conventions, are committed to …