A dual-agent framework leveraging left-brain logic (structured coding, syntax validation, debugging) and right-brain intuition (macro architecture, creative patterns, UX flows). It splits cognitive loads to boost LLM development efficiency, preventing hallucination while maximizing system scalability.

MCP for spawning sub-agents everywhere

Run DeepSeek as a real sub-agent inside Claude Code / Codex CLI — DeepSeek gets its own 7-tool agent loop in a sandboxed workspace, not just a single LLM call.

Spawn and manage sub-agents in visible tmux panes — the orchestrator that makes multi-agent boss mode work for pi coding agent.

agentic-dev-docs 是一个面向 Codex 的文档优先开发规划 skill。它不会直接替你开工写产品代码，而是先把项目想法、需求说明或开发任务整理成一套可以交给 Codex 新窗口执行的“主控开发控制包”。 这套控制包的目标是让后续开发过程更可控：先明确项目画像、阶段边界、权限矩阵、workstream 分工、Review 和审计规则，再由唯一的 A 主控智能体协调执行。

Quiet MCP executor for Claude Code and Codex. Offload edits, tests, refactors, and commits to cheaper local or flat-rate models.

Sage — Your Genie for OpenCode

AI Creative Production Sub-Agent Orchestration Framework — dispatch director, art director, cinematographer, brand director, and AI artist agents

Quality-first automatic model routing for Claude Code sub-agents. Default opus, downgrade only on confident trivial/routine tasks.

将 OpenCode 作为受管 MCP 子代理，支持任务追踪、会话复用、模型选择、权限处理和取消管理。

Open protocol & Cursor Skill for verifiable AI ID, agent lineage, capability ranking, cross-AI review, anti-Sybil consensus, and append-only final verdicts

Liquid Cluster Harness — 子 Agent 从用完即弃到自然生长的集群成员

Mine your Claude Code session logs into a personalized profile, twin sub-agent, and CLAUDE.md patch — local-first, your logs never leave your machine.

learning on agents related topics

Open-source Rust desktop AI coding agent — native egui GUI alternative to Claude Code / Cursor / Aider. 22-agent zero-trust validation autonomous loop MCP and Skills compatible zero WebView.