1010</div >
1111
1212## Table of Contents
13+
1314- [ Table of Contents] ( #table-of-contents )
1415- [ 1. Introduction] ( #1-introduction )
1516- [ 2. Features] ( #2-features )
@@ -104,6 +105,7 @@ python3 -m src.alertmanager_mcp_server.server --transport http --host 127.0.0.1
104105```
105106
106107Notes:
108+
107109- The ` stdio ` transport communicates over standard input/output and ignores host/port.
108110- The ` http ` (streamable HTTP) and ` sse ` transports are served via an ASGI app (uvicorn) so host/port are respected when using those transports.
109111
@@ -170,9 +172,12 @@ $ docker run -e ALERTMANAGER_URL=http://your-alertmanager:9093 \
170172 " run" ,
171173 " --rm" ,
172174 " -i" ,
173- " -e" , " ALERTMANAGER_URL" ,
174- " -e" , " ALERTMANAGER_USERNAME" ,
175- " -e" , " ALERTMANAGER_PASSWORD" ,
175+ " -e" ,
176+ " ALERTMANAGER_URL" ,
177+ " -e" ,
178+ " ALERTMANAGER_USERNAME" ,
179+ " -e" ,
180+ " ALERTMANAGER_PASSWORD" ,
176181 " ghcr.io/ntk148v/alertmanager-mcp-server:latest"
177182 ],
178183 "env" : {
@@ -215,6 +220,7 @@ The MCP server exposes tools for querying and managing Alertmanager, following [
215220### Pagination Benefits
216221
217222When working with environments that have many alerts, silences, or alert groups, the pagination feature helps:
223+
218224- ** Prevent context overflow** : By default, only 10 items are returned per request
219225- ** Efficient browsing** : LLMs can iterate through results using ` offset ` and ` count ` parameters
220226- ** Smart limits** : Maximum of 50 items per page prevents excessive context usage
@@ -238,8 +244,7 @@ $ make setup
238244# Run test
239245$ make test
240246# Run in development mode
241- $ mcp dev
242- $ TRANSPORT_MODE=sse mcp dev
247+ $ mcp dev src/alertmanager_mcp_server/server.py
243248
244249# Install in Claude Desktop
245250$ make install
0 commit comments