Feedback
- list_feedback
- get_feedback
- update_feedback
- resolve_feedback
Read the board with votes and comments, move requests along, and resolve them with a public reply.
Connect Claude Code, Cursor or any MCP client to FeedFast. Your assistant reads what users asked for, closes the request and publishes the changelog, without you leaving the editor.
Your assistant can read the repo, write the fix and deploy it. It can’t see what users asked for, or tell them it’s done. That’s the half this adds.
Your agent calls list_feedback and gets the real requests back, with votes and comments, from your real users.
This part is all yours. FeedFast never sees your code, your repo or your deploys. Your agent just knows which fix forty people voted for.
resolve_feedback marks it Done, replies with an Owner badge and publishes the changelog entry the reply links to.
schedule_maintenance puts it on your status page in blue, not red. No incidents, no false alerts, and uptime % skips the window.
get_uptime returns status, latency and open incidents. Worth asking right after a deploy.
Point your client at the hosted server, or run it locally through npx. Either way all it needs is an API key from Settings, Account.
Nothing to install and nothing to update: the server runs on ours. Works in Claude Code, Claude Desktop, Cursor, Windsurf and VS Code.
Claude Code
One command, then restart the session.
claude mcp add --transport http feedfast https://feedfa.st/api/mcp --header "Authorization: Bearer ff_your_key_here"Claude DesktopCursorWindsurfVS Code
The same object everywhere: only the config file's location differs.
{
"mcpServers": {
"feedfast": {
"type": "http",
"url": "https://feedfa.st/api/mcp",
"headers": {
"Authorization": "Bearer ff_your_key_here"
}
}
}
}The key comes from Settings → Account → API keys, and both transports take the same one. A read-only key lets an agent answer questions but never change anything. Give it a read-write key only when you want it to publish. The local form runs feedfast-mcp, fetched from npm on each run.
Read the board with votes and comments, move requests along, and resolve them with a public reply.
Write markdown release notes and publish them to your page, RSS feed and widgets at once.
Announce downtime ahead of time, then start, complete or cancel it.
Status, latency and uptime per monitor, plus any open incident and its cause.
The MCP server has no database connection. It calls the same public API you could call with curl, holding your key, so an agent can never do more than that key allows.
Reading works on every plan. Publishing, resolving and scheduling come with Pro, alongside five projects and three monitors in each. Three days free. The docs have every tool’s exact schema.
Cookies, kept honest
Cookies keep you signed in. We count visits anonymously unless you decline, and can remember your timezone if you let us. No advertising. Details