Version updated for https://github.com/0xmariowu/AgentLint to version v1.1.8. This action is used across all versions by 0 repositories. Action Type This is a Composite action. Go to the GitHub Marketplace to find the latest changes. Action Summary AgentLint is a GitHub Action designed to analyze and score the “harness” of AI agents—comprising configurations, rules, and supporting files that guide large language models (LLMs) in delivering accurate and effective results. It automates the evaluation process with 51 deterministic checks across key dimensions such as safety, workability, and continuity, and provides actionable fixes to optimize agent performance. By identifying and resolving issues in the harness, it helps prevent inefficiencies and AI-related errors, ensuring reliable and refined agent outputs.

Version updated for https://github.com/0xmariowu/AgentLint to version v1.1.7. This action is used across all versions by 0 repositories. Action Type This is a Composite action. Go to the GitHub Marketplace to find the latest changes. Action Summary AgentLint is a linter designed to evaluate and improve the “harness” that surrounds AI coding agents, ensuring reliable and high-quality performance. It automates the assessment of configuration files, CI setups, and instruction layers through 51 core deterministic checks and optional AI-driven extended checks, identifying and addressing issues that can lead to suboptimal agent behavior. By scoring and providing actionable fixes for the agent’s environment, it helps teams optimize their workflows and reduce errors in AI-assisted development.

Version updated for https://github.com/0xmariowu/AgentLint to version v1.1.6. This action is used across all versions by 0 repositories. Action Type This is a Composite action. Go to the GitHub Marketplace to find the latest changes. Action Summary AgentLint is a GitHub Action that evaluates and optimizes the “harness” components surrounding AI coding agents, such as configuration files, rules, and workflows, to improve their reliability and effectiveness. It automates the identification of issues across multiple dimensions, offering evidence-backed checks and actionable fixes to reduce errors, improve context clarity, and enhance overall agent performance. This tool addresses common challenges in harness engineering, ensuring AI agents deliver consistent and high-quality outputs.

Version updated for https://github.com/0xmariowu/AgentLint to version v1.1.5. This action is used across all versions by 0 repositories. Action Type This is a Composite action. Go to the GitHub Marketplace to find the latest changes. Action Summary AgentLint is a GitHub Action designed to evaluate and improve the “harness” components (e.g., configuration files, instructions, and CI setups) that support AI coding agents like Claude, Codex, and Cursor. By performing 51 deterministic checks across six core dimensions and offering optional AI-powered extended checks, it identifies and remediates errors or inefficiencies in the harness, ensuring agents deliver consistent, high-quality results. This action automates the process of scoring, diagnosing, and fixing harness issues, helping teams optimize their AI-assisted development workflows.