mcpField
B
Beamtrace

Beamtrace

VerifiedFeaturedRemote
beamtrace.comData & Analytics0Updated 7w ago

About Beamtrace

Beamtrace provides Cursor with remote MCP access to AI visibility data, including topic, prompt, and competitor analysis. Its read-only tools also surface content improvements for teams tracking brand visibility in AI search.

Key features

  1. 01Check overall visibility score
  2. 02Summarize topic performance
  3. 03Inspect prompt metrics and coverage gaps
  4. 04View competitor rankings
  5. 05Retrieve prioritized content improvements

Use cases

  1. 01Review a brand's overall AI search visibility
  2. 02Find gaps in prompt coverage
  3. 03Compare visibility against competitors
  4. 04Prioritize content improvements

MCP support

Connects to Beamtrace's remote MCP at https://beamtrace.com/api/mcp using OAuth. Its read-only tools expose AI visibility data to Cursor.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://beamtrace.com/api/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

5

This tool takes no parameters.

Details

From the project README

Beamtrace for Cursor

Cursor plugin that connects to Beamtrace's remote MCP for read-only AI visibility insights.

Also includes server.json for the MCP Registry (remote-only) and a stdio introspection shim for catalog Docker builds.

What's included

  • MCP: https://beamtrace.com/api/mcp (OAuth; organization's website is resolved automatically)
  • Rule: how to interpret metrics and phrase human reports

Tools

ToolPurpose
get_visibilityOverall visibility score and insight
list_topicsTopic rollups
list_promptsPrompt-level metrics and coverage gaps
list_competitorsVisibility leaderboard
list_improvementsPrioritized content improvements

Stdio catalog image

The product MCP is remote (https://beamtrace.com/api/mcp). This repo also ships a stdio shim that lists the same five tools so a Docker image can start and answer introspection (initialize / tools/list). The shim does not call Beamtrace APIs.

docker build -t beamtrace-mcp .
docker run -i --rm beamtrace-mcp

Validate

pnpm validate

Pre-commit runs the same check via Lefthook (pnpm exec lefthook install after clone).

Local install (dev)

ln -sf "$(pwd)" ~/.cursor/plugins/local/beamtrace

Reload Cursor → Customize → confirm Beamtrace loads → authenticate MCP.

Release

Pushing a v* tag (for example v1.0.1) runs .github/workflows/publish-mcp.yml. It logs in with HTTP domain auth, then:

node scripts/publish.mjs --target all --version "$GITHUB_REF_NAME" --yes

That sets .cursor-plugin/plugin.json and server.json to the tag version (v prefix is stripped), publishes to the MCP Registry, then commits those files to the default branch if they changed. The tag itself stays on the original commit.

Add repository secret MCP_PRIVATE_KEY: the 64-character hex private key used with mcp-publisher login http (not the PEM file).

git tag v1.0.1
git push origin v1.0.1

Local / interactive (install mcp-publisher first):

pnpm release

Pick cursor / mcp registry / all, then patch / minor / major, or pass --version. Cursor only bumps .cursor-plugin/plugin.json. MCP also runs mcp-publisher publish.

pnpm release -- --target cursor --bump patch --yes
pnpm release -- --target mcp --bump minor --yes
pnpm release -- --target all --bump patch --yes
pnpm release -- --target all --version 1.0.1 --yes

Namespace com.beamtrace/mcp needs domain auth for beamtrace.com (docs). Local login: pnpm registry:login:http. Use pnpm registry:login:github only for GitHub-owned namespaces.

Keep server.json version aligned with the live MCP serverInfo.version when you publish.

Auth

After install, complete Beamtrace sign-in when Cursor prompts. You need a Beamtrace account with a website at beamtrace.com/setup.

Publisher: Elfsight.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.