All Ears MCP server
Connect Claude, Cursor, ChatGPT and other AI assistants to All Ears media monitoring, so your assistant can read your saved monitors, pull metrics, break down sentiment and quote the actual mentions through natural conversation.
Contents
How do I set up the All Ears MCP server?
What do I need first?
- An All Ears account, with the Monitor API product on your plan; if you're not sure what's enabled, ask us (see below).
- An MCP-compatible AI client (Claude, Cursor, ChatGPT, or any custom agent).
https://mcp.allears.ai
Transport: Streamable HTTP. Authentication: OAuth (handled automatically
by your client — you just log in once in the browser).
How do I connect my AI client?
Claude (Desktop & Web — Custom Connectors). Go to Settings → Connectors → Add
custom connector, give it a name (e.g. All Ears), and paste the server URL
https://mcp.allears.ai. Claude will open a browser window for you to log in to
All Ears the first time you use a tool.
Claude Code:
claude mcp add --transport http allears https://mcp.allears.ai
Cursor. Open Settings → MCP → Add new MCP server, choose type Streamable
HTTP, and use the URL https://mcp.allears.ai. Or add this to your mcp.json:
{
"mcpServers": {
"allears": {
"url": "https://mcp.allears.ai"
}
}
}
ChatGPT. In a Developer mode / connectors-enabled workspace, add a new connector
pointing at https://mcp.allears.ai and complete the OAuth login when prompted.
Other clients. Any MCP client that supports remote servers over Streamable
HTTP with OAuth will work. Point it at https://mcp.allears.ai; most clients accept
a config block like the Cursor example above.
How do sign-in and permissions work?
The first time your assistant calls an All Ears tool, a browser window opens where you log in with your normal All Ears credentials and approve access. After that, your client manages tokens automatically. You'll be asked to grant these permissions:
| Permission | What it allows |
|---|---|
monitor:read |
List monitors, read mentions, metrics and analytics |
monitor:write |
Create, update and archive monitors |
Every All Ears organization can use the monitor tools over MCP; no extra product needs to be enabled. Read-only users cannot use the API.
What is a Monitor?
A Monitor is a saved query — keywords, sources and languages — whose
matching mentions are continuously collected and enriched with sentiment, reach
and PR value. Tools cover the lifecycle (list_monitors, get_monitor,
create_monitor, update_monitor, archive_monitor), the enriched mention feed and
daily metrics (get_mentions, get_metrics), and the monitor_* analytics
operators (counts, unique channels, trends, spikes, growth, top sources/terms, share
of voice, sentiment, and cursor-paginated export).
Entitlement: all monitor tools are available to every organization. The lifecycle and read tools need
monitor:read(andmonitor:writeto create, update or archive); themonitor_*analytics operators needmonitor:read.
What you can do with it: read the latest enriched mentions and daily metrics, size a monitor's coverage, chart its trend and flag spikes, measure period-over-period growth, rank the top channels (with sentiment per channel), split share of voice by channel type, break down sentiment, surface the conversation's over-represented terms, export every mention, and compare several monitors (e.g. you vs competitors) at once.
Try these prompts:
- "Show me all my monitors."
- "What were the latest mentions for my 'Climate Change' monitor?"
- "How did sentiment split for that monitor last week?"
- "Which podcasts drove the most coverage on my brand monitor in May, and how positive were they?"
- "Create a new monitor tracking 'electric vehicles' and 'EV' in English."
Full tool reference, example calls/responses, and the exact limits: see the Monitor use-case guide (agent-readable Markdown), and semantics & caveats for what the answers do (and don't) mean.
What's new?
See the changelog for the latest updates to the All Ears API and MCP tools.