Agentic Observability
by mdfifty50-boopcommunityv0.1.1
An MCP server for AI agent observability that logs agent actions, tracks token usage and costs, monitors tool performance, and detects anomalies like loops, latency spikes, and cost spikes. It also exposes static resources for LLM pricing and observability best practices.
From the registry: Agent tracing, cost tracking, anomaly detection for LLM agents
Install & configure
Prerequisites
- requiredInstall Node.js 18+Node.js version 18 or newer is required to run the MCP server via npx.
Install in your client
Paste into the AI you're chatting with — it will figure out the rest
Please install the `agentic-observability` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Node.js 18+** — Node.js version 18 or newer is required to run the MCP server via npx.
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["agentic-observability-mcp"]
Note: README shows the same stdio configuration for Claude Desktop, Claude Code, Cursor, and Windsurf/VS Code. No API keys or other environment variables are required. Server uses in-memory storage, so observability data is lost on restart.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Note: README shows the same stdio configuration for Claude Desktop, Claude Code, Cursor, and Windsurf/VS Code. No API keys or other environment variables are required. Server uses in-memory storage, so observability data is lost on restart.
Tools (6)
trace_agent_action
Log agent actions such as tool calls, LLM requests, decisions, and errors with metadata and timestamps.
track_token_usage
Track token usage for an LLM call and calculate its cost using built-in pricing tables.
get_cost_report
Return aggregated cost breakdowns across sessions grouped by model, provider, tool, or session.
log_tool_call
Log MCP tool calls with latency, success or failure status, and error details.
get_session_summary
Return a session report including cost, tokens, tool stats, error count, model breakdown, and duration.
detect_anomaly
Check for unusual patterns such as cost spikes, error-rate spikes, latency spikes, loop detection, and token explosions.
Prerequisites
- • Node.js 18+
Packages
agentic-observability-mcp
npmv0.1.1transport: stdio
Details
- Language
- JavaScript
- Released
- Apr 23, 2026
- Last commit
- Apr 24, 2026
- Registry name
- io.github.mdfifty50-boop/agentic-observability
Categories
Related Servers
Claude Flow
AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
64.6k
Codebase Memory Mcp
Persistent codebase knowledge graph. Survives session restarts and context compaction.
31.9k
Serena
A powerful toolkit for coding, providing semantic retrieval and editing capabilities.
26.5k