mayerwin

Agent Sentinel — AI skill for Claude Code

AI community

AI Cognitive Supervisor, Autonomous Resumption Engine & Rate Limit Sentinel for Claude Code Agents.

How to install Agent Sentinel

This entry records only its repository, not the path inside it, so there is no exact command to give. Open mayerwin/Agent-Sentinel and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Agent Sentinel does

AI Cognitive Supervisor, Autonomous Resumption Engine & Rate Limit Sentinel for Claude Code Agents.

Alternatives in AI

Ralph For Claude Code — by Frank Bria - An autonomous AI development framework that enables Claude Code to work iteratively on project 8k ★
J Space Cognition Suite V3.7 — J-Space Cognition Suite V3.7 - AI cognitive-enhancement Skills based on Anthropic's J-space global workspace r 3k ★
OpenSwarm — OpenSwarm — Autonomous AI dev team orchestrator powered by Claude Code CLI 589 ★

README

Agent Sentinel 🛡️🧠

**AI Cognitive Supervisor, Autonomous Resumption Engine & Rate Limit Sentinel for Claude Code Agents.**

[![Node.js Version](https://img.shields.io/badge/node-%3E%3D18.0.0-brightgreen.svg)](https://nodejs.org) [![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](https://opensource.org/licenses/MIT) [![Powered by Antigravity](https://img.shields.io/badge/Supervisor-Google%20Antigravity-8A2BE2.svg)](https://deepmind.google) [![Architecture: LLM Cognitive](https://img.shields.io/badge/Architecture-Cognitive%20Reasoning-orange.svg)](#how-it-works)

![Agent Sentinel Dashboard Overview](docs/screenshots/dashboard-overview.png)

Agent Sentinel is an intelligent supervisory dashboard and daemon designed for multi-agent Claude Code development workflows. Instead of relying on fragile regex matching, Agent Sentinel uses an **LLM Cognitive Supervisor** to analyze transcript semantics, distinguish between active execution and true rate limits, automatically issue resume instructions (*"continue"*), autonomously trigger bug-free hardening reviews upon task completion, and optionally hibernate your machine on weekly quota exhaustion.

🌟 Key Features

1. 🧠 LLM Cognitive Supervision (No Naive Regexes)

Evaluates raw conversation turns using continuous LLM reasoning.
Accurately differentiates between task completion, active tool execution, momentary network hiccups, 5-hour session ceilings, and weekly subscription limits.
Renders an explicit "🧠 LLM Supervisor Interpretation" card on each agent with step-by-step diagnostic reasoning.

2. 🤖 Three Autonomous Prompting Modes

Configure behavior globally in Settings or override per agent:

Auto Continue (continue) (Enabled by default):
- Automatically triggered when an agent reaches its quota/rate limit reset timestamp.
Auto Fix (The Ultimate Adversarial Bug-Free & Pixel-Perfect Hardening Loop):
- When an agent finishes its goal, Sentinel aut