AgentWatch
Supports UIby mishanefedov
Local observability and control plane for AI coding agents with a unified timeline, cost tracking, and anomaly detection.
What it does
Provides a local-first observability layer for multiple AI agents on a single machine. It aggregates events from various agents into a single timeline, tracks token usage/costs, and detects behavioral anomalies without any cloud telemetry.
Tools
list_recent_sessions: Retrieves a list of recent agent sessions across different projects.get_session_events: Pulls raw JSONL event streams for a specific session.search_sessions: Performs substring searches across all historical session files.get_tool_usage_stats: Aggregates tool call counts and durations per session.get_session_cost: Calculates the total USD cost of a session based on model rate tables.
Installation
Add to claude_desktop_config.json:
{
"mcpServers": {
"agentwatch": {
"command": "npx",
"args": ["-y", "@misha_misha/agentwatch", "mcp"]
}
}
}
Supported hosts
Claude Desktop, Cursor, Codex, Gemini CLI, OpenClaw.
Quick install
npm i -g @misha_misha/agentwatchInformation
- Pricing
- free
- Published
- 7/5/2026
- stars
- 0
Categories
Choose your AI client and follow the steps below.
Codex
Add to AGENTS.md or mcp configCursor
Add the above JSON to ~/.cursor/mcp.jsonGemini CLI
Add to settings.json mcpServers entryClaude Desktop
{
"mcpServers": {
"agentwatch": {
"command": "npx",
"args": ["-y", "@misha_misha/agentwatch", "mcp"]
}
}
}Related Apps
MCP TradingView Server
MCP ServerAccess TradingView technical indicators and market OHLCV data for automated trading analysis within AI agents.
Home Energy
MCP AppTrack household appliance energy usage with interactive simulation knobs and real-time MCP-driven controls.
Tuteliq MCP
MCP App41 AI-powered tools for child safety, fraud detection, and content moderation — bullying, grooming, sextortion, romance scams, and more — with interactive UI wi
Sprite MCP Server
MCP AppManage Sprite VMs with interactive dashboards, terminal UIs, and filesystem checkpointing.
Sentinel AIOps
MCP ServerAI-powered monitoring for ML models featuring real-time anomaly detection and drift classification.
AWS FinOps MCP Server
MCP AppAnalyze cloud costs, audit for waste, and monitor AWS budgets using natural language directly in your AI assistant.
Dynamic Model Autoscaling
MCP AppMetrics-based GPU autoscaling for LLM inference services on OpenShift AI using KEDA and vLLM.
RISKEN MCP Server
MCP ServerOfficial MCP server for RISKEN, enabling AI-driven security alert monitoring, finding analysis, and remediation automation.