PrometheusSAFE
A Model Context Protocol (MCP) server implementation that provides AI agents with programmatic access to Prometheus metrics via a unified interface.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Prometheus MCP Server
[](https://www.npmjs.com/package/prometheus-mcp) [](https://github.com/idanfishman/prometheus-mcp/pkgs/container/prometheus-mcp) [](https://codecov.io/gh/idanfishman/prometheus-mcp) [](https://nodejs.org/) [](https://opensource.org/licenses/MIT)
A Model Context Protocol (MCP) server that provides seamless integration between AI assistants and Prometheus, enabling natural language interactions with your monitoring infrastructure. This server allows for effortless querying, discovery, and analysis of metrics through Visual Studio Code, Cursor, Windsurf, Claude Desktop, and other MCP clients.
Key Features
- Fast and lightweight. Direct API integration with Prometheus, no complex parsing needed.
- LLM-friendly. Structured JSON responses optimized for AI assistant consumption.
- Configurable capabilities. Enable/disable tool categories based on your security and operational requirements.
- Dual transport support. Works with both stdio and HTTP transports for maximum compatibility.
Requirements
- Node.js 20.19.0 or newer
- Access to a Prometheus server
- VS Code, Cursor, Windsurf, Claude Desktop or any other MCP client
Getting Started
First, install the Prometheus MCP server with your client. A typical configuration looks like this:
6913cb0d7bdeOBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add prometheus-mcp -- npx -y [email protected]
{
"mcpServers": {
"prometheus-mcp": {
"command": "npx",
"args": [
"-y",
"[email protected]"
]
}
}
}Exposed tools (10)
8 read · 2 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
prometheus_build_info | read | Get Prometheus build information |
prometheus_label_values | read | Get available values for a specific label |
prometheus_list_labels | read | List all available Prometheus labels |
prometheus_list_metrics | read | List all available Prometheus metrics |
prometheus_list_targets | read | List all available Prometheus targets |
prometheus_metric_metadata | read | Get metadata for a specific Prometheus metric |
prometheus_query | write | Execute PromQL queries against Prometheus |
prometheus_query_range | write | Execute PromQL range queries for time-series data |
prometheus_runtime_info | read | Get Prometheus runtime information |
prometheus_scrape_pool_targets | read | Get targets for a specific scrape pool |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (2 observation(s))
- Network
- declared (1 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (6)
.prettierignore
import packageJSON from "../../package.json";
vi.mock("../../package.json", () => ({import packageJSON from "../../package.json";
@modelcontextprotocol/sdk, dotenv, express, pino, yargs, zod, @eslint/js, @tsconfig/recommended
You are a senior PostgreSQL DBA creating executive-level reporting for technical leadership. Your audience includes database administrators, application developers, infrastructure teams, and technical
Gates applied: no_behavioural_pass.
6913cb0d7bdefull audit observations/trust-audit/mcp-server/idanfishman__prometheus-5.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 6913cb0d7bde | SAFE | B | 89 | first audit |
Questions
What is the Prometheus MCP server?
A Model Context Protocol (MCP) server implementation that provides AI agents with programmatic access to Prometheus metrics via a unified interface.
What tools does Prometheus expose?
10 in total: 8 read-only, 2 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Prometheus safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does Prometheus need?
No credential environment variables were found in its source, so it appears to need none.
How does Prometheus run?
It speaks stdio and streamable-http, so it runs as a local process your client starts. It is published on npm as prometheus-mcp at 1.1.3.
How current is this page?
The grade is for one exact copy of the source (6913cb0d7bde), read on 2026-10-08. The repository is watched and re-audited when it changes.