Audio MCP runs on your own machine — the client starts it, so there is no endpoint to ping. 1 385 installs a week from pypi. Last commit 17 Oct 2026.
Audio perception for AI agents: measure, see, compare, edit and synthesize sound locally.
We read the source, 2 d ago · rules ccca72095570
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
proc = subprocess.run(cmd, stdin=subprocess.DEVNULL, stdout=subprocess.PIPE, stderr=err, timeout=timeout)
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
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 audio-mcp -- uvx audio-mcp
{
"mcpServers": {
"audio-mcp": {
"args": [
"audio-mcp"
],
"command": "uvx"
}
}
}
[mcp_servers.audio-mcp]
command = "uvx"
args = ["audio-mcp"]
{
"mcpServers": {
"audio-mcp": {
"args": [
"audio-mcp"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"audio-mcp": {
"args": [
"audio-mcp"
],
"command": "uvx"
}
}
}
This one needs environment variables set before it will start:
AUDIO_MCP_ALLOWED_DIRS (Optional sandbox: directories the server may read and write (separated by ';' on Windows, ':' elsewhere). Unset means unrestricted.), AUDIO_MCP_MAX_SECONDS (Optional: longest section (seconds) loaded into memory per call. Default 1800.).
The author declared them in the registry entry; get the values from the project itself.
Persistent memory for AI agents — log and recall conversation context over MCP.
One MCP server for every measured agent: search, compare, call agents paid per success, on a budget
Context selection and compression for AI agents - recall measured, not claimed.
Deterministic SVG perception for LLM agents — measures, never guesses. MCP server.
Local-first, auditable memory for MCP-compatible AI agents with explicit writes and SQLite storage.
Long-term and multimodal memory for AI agents - character-aware, mem0-compatible, fully-local option
Shared, permission-aware company context for AI agents, with provenance, approvals and audit.
Local visual UI cache for AI agents using perceptual hashing, CLIP, and 3D spatial grounding.
Answers built from our own checks of this server.