agent-workflows

中文优先的 Vibe Coding / AI coding engineering workflow guide: specs, agents, worktrees, skills, CI, and review.

Re-anchor long conversations that drift through repeated local corrections.

Pressure-test research claims with falsifiable evidence plans, adversarial checks, frozen verifiers, and proof ledgers.

AI agent tools for coding assistants: skills, MCP servers, agent loops, subagents, hooks, plugins, and CLI tools. Machine-readable catalogs. Works with Claude Code, Cursor, Codex, Gemini CLI, and 30+ agents.

Visual session-state signals for OpenCode using its native theme system

Readable agent skills with optional declarative YON protocols you can inspect and validate. A personal, field-used pack by Alexandru Mares.

Verified workflow engine for Claude Code: durable project memory, phase planning, subagents, and deterministic completion gates.

Agentic engineering across machines, under control.

Catch when your agent skills, instructions, and rules stop triggering — or stop working.

Visual debugging, tracing, and replay for agent workflows.

A repository-native task coordination framework for coding agents.

Chinese AI Agent skills for goal definition and loop engineering

Shared workflow skills for agent-driven engineering across repositories.

Capture what matters without derailing the conversation — a shared attention workspace for Codex & Work in ChatGPT Desktop.

Multi-agent orchestration for Orca, as code — LLM-authored plans that fan out worker agents.

Codex-first GSD: spec-driven development, context engineering, and verified AI coding workflows for OpenAI Codex.

Three AI agents that check each other — Planner, Generator, Evaluator with contract negotiation. Multi-agent dev workflow skill for WorkBuddy & Claude Code. 三个 AI 互相独立验收的开发工作流 skill，支持 WorkBuddy 与 Claude Code。

Agent Skill for AI-native startup stage gates, MVP planning, PMF diagnostics, launch operations, and scale-stage moat workflows.

Sanitized CuraOS multi-agent workspace example

Governed operating loops and portable capability Capsules for persistent AI agents