MCP Audit runs on your own machine — the client starts it, so there is no endpoint to ping. 55 installs a week from npm. Last commit 14 May 2026.
Free context-engineering audits for AI agents. BYOK Anthropic key. Top-3 findings per scan.
Nothing serious here today
Today is the operative word: we check MCP Audit every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
Connect this server
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add mcp-audit -e 'ANTHROPIC_API_KEY=<ANTHROPIC_API_KEY>' -- npx -y @archonics/mcp-audit{
"mcpServers": {
"mcp-audit": {
"command": "npx",
"args": [
"-y",
"@archonics/mcp-audit"
],
"env": {
"ANTHROPIC_API_KEY": "<ANTHROPIC_API_KEY>"
}
}
}
}[mcp_servers.mcp-audit]
command = "npx"
args = ["-y", "@archonics/mcp-audit"]
env = { ANTHROPIC_API_KEY = "<ANTHROPIC_API_KEY>" }{
"mcpServers": {
"mcp-audit": {
"command": "npx",
"args": [
"-y",
"@archonics/mcp-audit"
],
"env": {
"ANTHROPIC_API_KEY": "<ANTHROPIC_API_KEY>"
}
}
}
}{
"servers": {
"mcp-audit": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@archonics/mcp-audit"
],
"env": {
"ANTHROPIC_API_KEY": "<ANTHROPIC_API_KEY>"
}
}
}
}Put your own value in place of
<ANTHROPIC_API_KEY> (Your Anthropic API key. Required — the audit engine calls Claude directly from your machine).
The author’s registry entry marks it as required.
Alternatives to MCP Audit
same job, measured the same wayLocal context and engineering evidence for AI coding agents.
Temporal engineering context for agents via MCP.
Local-first persistent context memory MCP server for AI coding agents.
Memory, docs, and repo context layer for engineering agents.
Context engineering for AI coding agents: product context, project missions, and 360 memory.
Cache-aware context engine for AI agents: server-side payload filtering, honest context ledger.
Context Assurance for AI agents: context engineering, AI cost optimization, verification.
Memi is the read-only design engineering audit and skill layer for coding agents.
MCP Audit — questions
Answers built from our own checks of this server.