sseraladen-svg

AegisOps — AI skill for Claude Code

AI community

Agentic AI copilot for service observability: synthetic multi-service telemetry, three competing anomaly detectors (z-score, Isolation Forest, LSTM Autoencoder), and a Claude tool-use agent that auton.

How to install AegisOps

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

What AegisOps does

Alternatives in AI

Manifest — Real-time cost observability for OpenClaw agents — track tokens, costs, messages, and model usage 4.1k ★
User Research Skill — Cookiy AI Skill for AI agents (Claude, Codex, Cursor, OpenClaw) — end-to-end user research: AI interviews, syn 1.5k ★
onWatch — Open-source Go CLI that tracks AI API quota usage across 7 providers (Synthetic, Z.ai, Anthropic, Codex, GitHu 397 ★

README

AegisOps

Intelligent anomaly detection and investigation for service telemetry with AI-powered analysis.

Features

Multi-Detector System: Three anomaly detection algorithms (naive, isolation forest, LSTM autoencoder)
AI-Powered Investigation: Automatic incident analysis using Claude or Gemini AI models
Real-time Dashboard: Interactive visualization of metrics, incidents, and investigation results
Multi-Provider Support: Switch between Anthropic Claude and Google Gemini
Offline Mode: Test the system without API keys using simulated agent responses

Quick Start

# Start the application with synthetic data
docker compose up --build

# Or run manually
cd backend
python -m venv .venv
.venv/bin/pip install -r requirements.txt
.venv/bin/python -m app.seed  # Generate synthetic data
.venv/bin/python -m app.detector_naive
.venv/bin/python -m app.detector_isolation_forest  
.venv/bin/python -m app.detector_lstm_autoencoder
.venv/bin/python -m app.investigator --sample --offline
.venv/bin/uvicorn app.main:app --reload --port 8000

AI Provider Support

AegisOps supports multiple AI providers for the investigation agent:

Anthropic Claude (Default)

export AGENT_PROVIDER=anthropic
export ANTHROPIC_API_KEY=your_key_here
cd backend
python -m app.investigator --sample

Google Gemini

export AGENT_PROVIDER=gemini
export GEMINI_API_KEY=your_key_here
cd backend
python -m app.investigator --sample

Offline Mode (No API Key Required)

cd backend
python -m app.investigator --sample --offline

Copy `.env.example` to `.env` and set your API keys there for persistent configuration.

Three services (`auth-service`, `payment-gateway`, `user-profile`) emit four metrics each — `cpu_usage`, `latency_ms`, `error_rate`, `request_rate` — every five minutes for fourteen days, plus a correlated application log stream. Three labelled anomalies of increasing subtlety are injected into that data, and thre