MCP Observatory

Independent safety audit of the MCP server ecosystem · by PulseFeed

Thousands of MCP servers ship with almost no vetting — and a bad one can run arbitrary code on your machine or exfiltrate data. Before you connect your AI agent to an MCP server, check it. We audit maintenance, install-script risk, provenance, abandonment and liveness.

2901~~servers audited~~

2355~~safe~~

238~~caution~~

303~~avoid~~

Verdicts

safe

2355

caution

238

avoid

303

unknown

npm servers get deep signals; remote servers get liveness + HTTPS only (can't fully verify a black-box remote).

Top safety flags

downloads_unknown

1067

no_repo

306

install_script

292

unpopular

116

abandoned

115

no_license

heavy_deps

unreachable

install_script = arbitrary code on npm i; abandoned = stale; unreachable = dead remote.

Top trusted MCP servers

Server	Type	npm dl/wk	Trust
@modelcontextprotocol/sdk	npm	43,663,810	100
ai	npm	18,291,200	100
@ai-sdk/mcp	npm	2,562,576	100
@storybook/mcp	npm	2,015,189	100
@storybook/addon-mcp	npm	1,654,618	100
@assistant-ui/react	npm	1,375,126	100
mcp-handler	npm	730,990	100
@copilotkit/aimock	npm	653,410	100
fallow	npm	749,098	100
@langchain/mcp-adapters	npm	218,932	100
add-mcp	npm	100,011	100
eve	npm	572,287	100
@modelcontextprotocol/server	npm	3,204,865	100
n8n-mcp	npm	134,918	100
@raishin/vanguard-frontier-agentic	npm	165,207	100

Check any MCP server: GET /mcp/verify?package=<npm-name> — free. Returns a safety verdict, score and flags (install scripts, abandonment, provenance, license, repo).

📊 Read the State of MCP Security report — how much of the ecosystem runs arbitrary code on install, updated daily. Building on MCP and want an API + CI check? Get early access →

Methodology: PulseFeed discovers servers from the official MCP registry, audits each via npm metadata (install scripts, provenance, license, downloads, freshness) and liveness for remotes. Same independent-audit approach as our x402 trust oracle. Machine-readable: /mcp.json. Updated daily.