Point Claude Desktop, Claude Code, Cursor or any Model Context Protocol client at your workspace and it can search, read, create, update and comment on issues — as a real person, with that person's exact permissions. Every read goes through the same visibility rules the web interface uses and every write through the same workflow and field rules, with the owner of the API key as the actor. The feature adds one capability: whether you may connect a client at all. It never widens what is behind it.
Client setup is covered separately in connecting an MCP client.
On your own Redmine
Requirements
- Redmine 6.0 or newer (Rails 7.2, Ruby 3.1 or newer)
- PostgreSQL 14+, or MySQL 8.0+ / MariaDB 10.6+
- No additional gems, and no tables of its own
- HTTPS in front of Redmine — an API key travels on every request
Install
cd /path/to/redmine
cp -r redminepro_mcp plugins/
# no migrations — this feature owns no tables
# restart Redmine
Then name somebody under Who may connect and hand them the endpoint address.
Standalone behaviour
The plan gate opens. With no RedminePRO plan signal at all, the endpoint is available: there is no tier to enforce on hardware RedminePRO does not run, and refusing there would be billing a plan you never bought. The settings page says which case applies to your workspace.
The suite tools disappear rather than break — a tool whose feature is not installed is absent from the list and refused if called. Everything else is identical.
Update and remove
Copy the new directory over the old one and restart. To remove:
rm -rf plugins/redminepro_mcp
# restart Redmine
Nothing is lost: every change it made was an ordinary change to an issue, made by a real person, and it is all still in the history.
Already on RedminePRO Cloud? It is installed and maintained for you: see the configuration guide.