DeployAngel — Claude Code MCP server: what it does, how to install
Monitors production rollouts and feeds failure diagnostics directly to your coding agent.
claude mcp add deployangel -e DEPLOYANGEL_API_TOKEN=<DEPLOYANGEL_API_TOKEN> -- uvx deployangelWhat DeployAngel does
The MCP integration links your coding assistant with the DeployAngel monitoring service. The runtime agent collects metrics once per minute from web processes: HTTP status distributions, route latency histograms, unhandled exceptions, and background task queues such as Celery or RQ. This telemetry helps the platform determine whether a new version is cleared or failing.
Sensitive information remains private. The library excludes request bodies, URL arguments, headers, cookies, SQL queries, and logs, sending sanitized traces and route patterns instead. Telemetry streams via a background thread to hosted infrastructure on DigitalOcean in the US. If the remote collector becomes unreachable, an in-memory buffer holds up to 10 compressed packages before dropping older records.
You need a valid DEPLOYANGEL_API_TOKEN to run the tool. For production monitoring, the companion package expects Python 3.10 or higher alongside Django 4.2+, FastAPI 0.100+, Starlette 0.27+, or Flask 2.3+.
What it helps with
- verify whether a production release passed all stability checks
- inspect exception traces and error spikes after a deployment
- track stuck background workers across Celery or RQ queues
- map runtime failures to specific commit revisions
Example prompts for the agent
- Check if production has cleared our latest release
- Retrieve unhandled exceptions reported for the current FastAPI revision
- Inspect failed Celery jobs and queue latency since the last deploy
How to install DeployAngel
Platform CLI
claude mcp add deployangel -e DEPLOYANGEL_API_TOKEN=<DEPLOYANGEL_API_TOKEN> -- uvx deployangelReplace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from deployangel <command>
Platform CLI
codex mcp add deployangel --env DEPLOYANGEL_API_TOKEN=<DEPLOYANGEL_API_TOKEN> -- uvx deployangelReplace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from deployangel <command>
Config file
{
"mcpServers": {
"deployangel": {
"command": "uvx",
"args": [
"deployangel"
],
"env": {
"DEPLOYANGEL_API_TOKEN": "<DEPLOYANGEL_API_TOKEN>"
}
}
}
}Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from deployangel <command>
Config file
{
"servers": {
"deployangel": {
"type": "stdio",
"command": "uvx",
"args": [
"deployangel"
],
"env": {
"DEPLOYANGEL_API_TOKEN": "<DEPLOYANGEL_API_TOKEN>"
}
}
}
}Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from deployangel <command>
Platform CLI
gemini mcp add -e DEPLOYANGEL_API_TOKEN=<DEPLOYANGEL_API_TOKEN> deployangel uvx deployangelReplace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from deployangel <command>
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about DeployAngel
How to install DeployAngel in Claude Code?
Run claude mcp add deployangel -e DEPLOYANGEL_API_TOKEN=<DEPLOYANGEL_API_TOKEN> -- uvx deployangel in your terminal.
What does DeployAngel do?
It tracks application behavior across production processes and supplies your agent with aggregated diagnostic evidence when a release runs into trouble.
Does DeployAngel capture request payloads or raw customer data?
No, it ignores HTTP request bodies, cookies, database queries, and raw parameters, capturing only framework route definitions, latency metrics, and masked exception lines.
