Does SignlOS have an MCP server?

Yes. SignlOS exposes a Model Context Protocol endpoint at https://api.signlos.com/mcp, using streamable HTTP. It authenticates with the same scoped API key as the REST API, and a key's scopes limit which tools the assistant can call.

How do I connect Claude to my product feedback?

Create a SignlOS API key, then add https://api.signlos.com/mcp as a remote MCP server in Claude with that key as a bearer token. Claude can then read your boards, posts, roadmap, changelog, and status, and — if the key has write scopes — create posts and publish release notes.

Is it safe to give an AI assistant access to my feedback data?

The assistant can do exactly what the API key allows and nothing more. Issue a read-only key for questions and analysis; grant write scopes only if you want the assistant to create posts or publish releases, and revoke the key to revoke the assistant's access instantly.

Connect Claude to SignlOS

Point Claude, or any MCP client, at your workspace so you can ask what your customers are asking for — and act on the answer.

5 min read

Reading a feedback board is a job for a machine. There are four hundred posts, the same request is phrased six ways, and the question you actually have — *what are enterprise customers blocked on this quarter* — is not a filter you can express in a UI.

SignlOS speaks Model Context Protocol, the open standard for connecting AI assistants to real systems. Point Claude at your workspace and you can ask that question in English, and get an answer grounded in your actual posts and votes rather than a guess.

What you can ask

*What are the top ten requests by votes that we have not planned yet?*
*Summarise everything customers said about billing in the last 90 days.*
*Which roadmap items have shipped but have no release note?*
*Are there duplicates of this post I should merge?*
*Draft a release note for the items we moved to Shipped this week.*

Connect it

1
Create an API key.
Settings → API keys → New key. For asking questions, grant only the :read scopes. Copy the key — it is shown once.
2
Add SignlOS as a remote MCP server.
In Claude, add a custom connector with the URL https://api.signlos.com/mcp and your key as the bearer token. Claude Desktop, Claude Code, and any other MCP client that speaks streamable HTTP all connect the same way.
3
Ask something.
*What are our most requested features this month?* If the connection is live, the answer cites real posts and vote counts.

{
  "mcpServers": {
    "signlos": {
      "type": "http",
      "url": "https://api.signlos.com/mcp",
      "headers": {
        "Authorization": "Bearer sk_live_xxxxxxxxxxxxxxxxxxxx"
      }
    }
  }
}

Claude Desktop — claude_desktop_config.json

claude mcp add --transport http signlos https://api.signlos.com/mcp \
  --header "Authorization: Bearer sk_live_xxxxxxxxxxxxxxxxxxxx"

Claude Code — one command

The tools

Tool	What it does	Scope needed
`list_boards`	Every board in the workspace, with post counts.	`boards:read`
`search_posts`	Search and filter posts by text, board, status, tag, or votes.	`posts:read`
`get_post`	One post in full — description, votes, requesters, comments.	`posts:read`
`feedback_summary`	What the board says right now: top requests, movement, themes.	`posts:read`
`create_post`	File a new post on a board.	`posts:write`
`update_post_status`	Move a post, notifying everyone who voted for it.	`posts:write`
`get_roadmap`	Roadmap items by column, with the posts feeding each.	`roadmap:read`
`list_changelog`	Published release notes, newest first.	`changelog:read`
`get_status`	Component health and any open incidents.	`status:read`

What it can and cannot do

The MCP endpoint uses the same key, the same scopes, and the same tenant isolation as the REST API. An assistant holding a read-only key can read; it cannot write, whatever it is asked to do. There is no separate MCP credential to issue, audit, or forget about — revoking the API key revokes the assistant's access in the same moment.

Writes notify real customers

update_post_status emails everyone who voted on the post. It is not a reversible sandbox action. Start with a read-only key, and only add write scopes once you want the assistant acting on your behalf in public.

Stateless transport

Streamable HTTP (stateless). The API runs several tasks behind a load balancer, so nothing is held between calls — any state kept in one process would be a coin flip on the next request.

NextStatus badges and public data

Common questions

Does SignlOS have an MCP server?: Yes. SignlOS exposes a Model Context Protocol endpoint at https://api.signlos.com/mcp, using streamable HTTP. It authenticates with the same scoped API key as the REST API, and a key's scopes limit which tools the assistant can call.
How do I connect Claude to my product feedback?: Create a SignlOS API key, then add https://api.signlos.com/mcp as a remote MCP server in Claude with that key as a bearer token. Claude can then read your boards, posts, roadmap, changelog, and status, and — if the key has write scopes — create posts and publish release notes.
Is it safe to give an AI assistant access to my feedback data?: The assistant can do exactly what the API key allows and nothing more. Issue a read-only key for questions and analysis; grant write scopes only if you want the assistant to create posts or publish releases, and revoke the key to revoke the assistant's access instantly.

Something unclear or wrong? Tell us — or post it on our board.

Connect Claude to SignlOS

What you can ask#

Connect it#

The tools#

What it can and cannot do#

Common questions

What you can ask

Connect it

The tools

What it can and cannot do