ShunsukeHayashi/agent-skill-bus
Self-improving task orchestration framework for AI agent systems. DAG-based task queue + automatic skill quality monitoring + external change detection. Framework-agnostic (OpenClaw, Claude Code, Codex, LangGraph, CrewAI).
Implements a three-module closed-loop system for operational AI agent health: a JSONL-based task queue with DAG dependency resolution and file-level locking (Prompt Request Bus), an automated 7-step quality repair cycle triggered by score drops and drift detection (Self-Improving Skills), and external change monitoring that routes API/dependency/config updates into improvement workflows (Knowledge Watcher). Zero-dependency Node.js runtime that integrates via simple CLI commands and works standalone with any agent framework through skill execution logging.
Used by 1 other package. Available on npm.
Stars
12
Forks
3
Language
JavaScript
License
MIT
Category
Last pushed
Mar 18, 2026
Commits (30d)
0
Reverse dependents
1
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/ShunsukeHayashi/agent-skill-bus"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related agents
kill136/claude-code-open
Open source AI coding platform with Web IDE, multi-agent system, 37+ tools, MCP protocol. MIT licensed.
context-machine-lab/sleepless-agent
🤖 24/7 AI agent that maximizes Claude Code Pro usage via Slack. Auto-processes tasks, manages...
m0n0x41d/quint-code
Structured reasoning framework for Claude Code, Gemini, Cursor, and Codex — hypothesis-driven...
GobbyAI/gobby
A local-first daemon to unify your AI coding tools. Session tracking and handoffs across Claude...
chadbyte/clay
Turn Claude Code into a team workspace. Any device, one command.