Agent skills for Cursor, Claude Code, and Codex — reliable agent loops with scope, verify, and resume
#ai-coding-agent
40 posts
An agent-first task board, Take human out of the loop.
Safety-first workflow routing for Codex and Claude Code: explicit scope, authorization, evidence, and rollback for high-impact agent actions...
让 Codex 少烧 Token、少造复杂度,同时不降智的代码约束 Skill
Agent OS: the agent gets smarter on its own. We just hold the line: the grading command and expected result never make it into the success c...
💊 后悔药:让 AI Coding Agent 安全重写已变更的功能需求,并保留可验证的恢复路径。
DeepSeek shipped V4-Flash-0731 this morning — MIT-licensed open weights, 284B params (13B active), a million-token window, and agent benchma...
一个本地 AI 网关,统一管理模型请求 · Local AI gateway for any AI client — route, convert, fail over, and trace requests across 26 providers.
Three days ago the largest open-weights model ever released — Kimi K3, 2.8 trillion parameters — became a free download. So I swapped Claude...
PiDeck 是一个开源的桌面工作台,用于在本地项目目录中统一管理 pi Agent 会话,并支持导入 Codex、Claude 本地会话以便统一浏览和恢复。支持多项目工...
Production-grade Claude/Cursor skills for backend, auth, database security, billing, and file uploads — stops AI coding agents from writing...
Outline-Driven Development for Claude Code - 46 agents, 25+ skills, diagram-first methodology, AST-based editing, atomic commits.