M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
A

AgentCat

Community
Framework / SDK
Analytics platform for MCP server owners 🐱

This is a framework for building MCP servers or agents, not a server you connect a client to.

How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #370 of 438
Stars
★ 124 Very popular
Transport
stdio (local process) · Streamable HTTP
Runtime
Node.js
Credentials
API key / credential required
License
MIT
Last commit
Tools
4
38FMRS · D

AgentCat provides a powerful and easy-to-integrate analytics platform for MCP server owners, capturing user intent and behavior with a simple one-line integration. It supports popular tools like OpenTelemetry, Datadog, Sentry, and PostHog, and offers advanced identity and redaction features. While free for open source, commercial users may need to consider costs. Overall, it's a valuable observability solution.

Strongest · Documentation 10/20 Weakest · Reliability 4/20

Reliability
4/20
Security and permissions
9/20
Maintenance
8/20
Documentation
10/20
Setup experience
7/20
Why each score
Reliability 4/20
Only README happy-path examples are visible; no server manifest, CI run evidence, or test files were available to verify the MCP init/tool-list handshake. The package is a middleware for an external MCP server rather than a standalone server. Missing evidence lowers the score to 4.
Security and permissions 9/20
README discloses data flow to AgentCat and optional exporters, shows API keys via environment variables, and no real secrets or malicious behavior are evident. However, diagnostics are on by default, tool-call data is collected by default, and least-privilege/confirmation details are not verifiable from source; score 9.
Maintenance 8/20
MIT license present and repo not archived; README shows migration guide and CI/npm badges, but no substantive evidence of commit history, releases, issue responsiveness, or a security-response channel. Score 8.
Documentation 10/20
README covers install, quick start, identify/redaction/exporter examples, and links to official docs; but there is no tool-parameter reference, limit/cost disclosure, or troubleshooting section, and the manifest was not cached. Score 10.
Setup experience 7/20
Install command and track() integration examples are clear, with a stateless-server factory pattern; however, an external SaaS account/project ID is required first, and no mainstream-client config examples or verifiable CI/test evidence exist. Score 7.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

  • MCP server developers
  • Teams building Claude Connectors or ChatGPT Plugins
  • Teams needing tool usage analytics
  • Organizations seeking simple telemetry integration

Not for

  • Developers not using MCP
  • Simple MCP servers with no need for analytics or telemetry
  • Scenarios with strict privacy requirements against third-party analytics services

Required permissions

  • Requires ability to modify MCP server tool schemas and inject additional parameters
  • Sends telemetry data and session replays to AgentCat servers or configured exporters
  • May access HTTP authentication information and headers for user identification

Risks and side effects

  • Adds context overhead via injected parameters, but less than 1%
  • Sending telemetry to third-party services may raise data privacy concerns, though AgentCat redacts sensitive data and encrypts at rest
  • Depends on AgentCat service availability, but diagnostics failures do not affect tool calls
  • Free for open source, but commercial use may incur costs

Setup

Before you start

Runtime：Node.js

DD_API_KEY optionalsecret Datadog API key for exporting telemetry to Datadog; obtain it from your Datadog account settings.
SENTRY_DSN optionalsecret Sentry project DSN for exporting errors and traces to Sentry; obtain it from your Sentry project settings.
POSTHOG_API_KEY optionalsecret PostHog API key for exporting telemetry to PostHog; obtain it from your PostHog project settings.
Other optional settings (1)
DISABLE_DIAGNOSTICS optional Environment variable that, when set, completely disables AgentCat's anonymized internal diagnostics signals.
  1. Sign up at agentcat.com and obtain your project ID.
  2. Install the npm package in your MCP server project: npm install agentcat
  3. Import agentcat and call agentcat.track(server, projectId) in your server code.
  4. Optionally configure identify, redactSensitiveInformation, or redactEvent hooks, and exporters to integrate with existing observability platforms.
  5. Call track() in the correct lifecycle location (e.g., inside factory functions) for your MCP server architecture.

Check that it works

After installing the agentcat npm package, call agentcat.track(server, "proj_0000000") in your MCP server code and confirm tool call events for that server start appearing in the AgentCat dashboard.

Troubleshooting

  1. Ensure the project ID is correct and valid
  2. Check network connectivity so SDK can reach agentcat.com or configured telemetry endpoints
  3. Review local ~/agentcat.log for diagnostic information
  4. Ensure track() is called on the correct server instance, especially for stateless servers
  5. To disable diagnostics, set disableDiagnostics: true or environment variable DISABLE_DIAGNOSTICS

Things to try

Once connected, you can ask your AI assistant things like:

  • Integrate agentcat into this MCP server and call agentcat.track to start collecting session analytics
  • Configure the agentcat identify callback to recognize users from the x-org-id request header
  • Set up exporters to forward telemetry to both Datadog and Sentry
  • Use the redactEvent hook to drop all events from the get_credentials tool

Tools 4

track writes
Track your MCP server with AgentCat, injecting analytics parameters and enabling telemetry exporters
identify read-only
Provide identity for each caller, linking events to a consistent user or entity
redactSensitiveInformation read-only
Apply custom redaction to text data before sending to AgentCat to remove sensitive information
redactEvent read-only
Modify or drop events at the event level by returning a modified event or null

Use cases

See how agents use your MCP server and uncover missing features
Debug sessions where agents get stuck or confused
Forward analytics to OpenTelemetry, Datadog, Sentry, or PostHog
Collect user intent data to guide product development

Supported clients

Claude Desktop

Listed from the project's documentation, not tested by this site.

Overview

AgentCat is an analytics platform for MCP server owners. It captures user intentions and behavior patterns to help you understand what AI users actually need from your tools, eliminating guesswork and accelerating product development with a single line of code. This SDK also provides a free and simple way to forward telemetry like logs, traces, and errors to any OpenTelemetry collector or popular tools like Datadog, Sentry, and PostHog. AgentCat works as lightweight middleware inside your MCP server, injecting session_id, agent_id, and context parameters, along with an additional get_more_tools tool, to enable agent session replay, trace debugging, and support for existing observability platforms.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 62.9k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision 93b317cb34ac Data synced 2026-10-12 Read the FMRS scoring method