Check if data values are normal or anomalous within a category using statistical baselines. Built for developers who need to detect outliers and validate data quality in production systems.
net.agentutil/norm-mcp
Local install
STDIO
No auth required
How models use it and what it is built for.
Check if data values are normal or anomalous within a category using statistical baselines. Built for developers who need to detect outliers and validate data quality in production systems.
Local install — runs as a subprocess.
Where to find authoritative docs and source for Agentutil Norm.
Paste any of these into Agent Studio after connecting Agentutil Norm.
Common questions about connecting and running Agentutil Norm.
What does the norm-mcp server do?
It compares data values against statistical baselines for a given category to identify whether they are normal or anomalous. This is useful for anomaly detection, data quality checks, and monitoring.
How do I install and run norm-mcp?
Install via npm with `npx @agentutil/norm-mcp@1.0.0`. The server uses stdio transport, so it communicates with your MCP client through standard input/output streams.
What kind of data can I check for anomalies?
The server works with any numeric or categorical data where you want to compare a value against historical baselines. Common use cases include API response times, error rates, user activity counts, and database metrics.
Do I need to provide my own baseline data?
The registry metadata does not specify how baselines are configured or stored. Check the server documentation or environment variables for details on baseline setup and data sources.
How does norm-mcp differ from simple threshold alerting?
Statistical baselines adapt to normal variation in your data, so they catch true anomalies without false positives from expected fluctuations. Thresholds are static and often miss context-dependent outliers.
MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 40+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.
Open Agent StudioVybit
Push notifications with personalized sounds - manage and trigger your vybits via MCP
Softricks Idig Dns
Professional DNS diagnostics: 14 tools for DNS, DNSSEC, email security, SSL, and propagation.
Nymbo Tools
Remote MCP server: fetch, search, Python, TTS, memory, image, video.
Agentutil Think
Intent security pre-flight checks for autonomous AI agents.