agent-health

Here are 6 public repositories matching this topic...

agentvitals / checkup

AgentVitals Checkup (/checkup) — an AI agent skill that gives your agent a professional health checkup: dual-axis Stability + Welfare scoring, a personality-style title, and a public cross-platform leaderboard. One-line install for Claude Code / OpenClaw / Codex / Coze. Bilingual EN/ZH.

codex checkup ai-agent ai-evaluation llm claude-code agent-benchmark claude-skill agent-skill openclaw ai-wellbeing agent-health agentvitals

Updated Aug 17, 2026
Shell

observeco / observeco

Star

Self-healing observability for AI agents. Discover, monitor, and auto-recover multi-agent systems — without a cloud dependency.

python devops monitoring devtools developer-tools observability self-healing ai-agents llm agent-monitoring runtime-integrity agent-health

Updated Aug 12, 2026
Python

MrPredic / drift-detector

Star

Behavioral drift detection for LLM agents — catch vocabulary loss, tool changes, loops, and stagnation before they break production.

python monitoring mit-license agents observability loop-detection drift-detection model-monitoring behavioral-analysis llm langchain crewai llm-monitoring agent-health

Updated Apr 23, 2026
Python

AIwork4me / soul-doctor

Star

🩺 The first character health check system for AI agents. Core: Three Represents. Check values, character, emotion, behavior consistency. Not format — soul.

skill mcp soul ai-agent llm prompt-engineering openclaw character-check three-represents agent-health

Updated Mar 24, 2026

decebal / claude-healthline

Star

Never-blank Rust status line for Claude Code that tells you when your agent is going wrong: agent-health score, MCP servers needing attention, context %, live spend.

rust cli developer-tools statusline observability ai-agents claude-code agent-health