mcpField
S
Serpel

Serpel

Remote
serpel.appBusiness & Marketing0Updated 3w ago

About Serpel

Serpel connects AI assistants to account-specific SEO data, including rankings, search performance, audits, competitors and AI visibility. It is for teams monitoring their websites; its read-only tools also surface prioritized recommendations and keyword opportunities.

Key features

  1. 01Project health and audit summaries
  2. 02Keyword rankings and opportunities
  3. 03Search performance comparisons
  4. 04Competitor and AI visibility data
  5. 05Crawl issues and recommendations
  6. 06Conversion and codebase reports

Use cases

  1. 01Check keyword movements between rank checks
  2. 02Find queries ranking in positions 4–20
  3. 03Review crawl issues by severity
  4. 04Compare landing-page sessions and conversions

MCP support

The hosted MCP server exposes read-only tools for Serpel project data to connected AI clients. It uses Streamable HTTP at `https://serpel.app/mcp`; clients can sign in with OAuth or use an API token.

Configuration

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

Endpointhttps://serpel.app/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

Details

From the project README

Serpel plugin for AI assistants

Ask Claude, ChatGPT, Codex, Cursor, GitHub Copilot, VS Code or the Gemini CLI about the SEO data of your own websites. The plugin connects your assistant to the Serpel MCP server and adds a skill that explains which tool to use when.

The plugin contains manifests and one skill only. It has no program code and no hooks.

What you can ask

  • Which projects do I have, and how healthy is each one?
  • Which keywords moved up or down since the last rank check?
  • What do Google Search Console clicks and impressions look like compared with the previous period?
  • Which queries rank on positions 4 to 20 and are quick wins?
  • Where do my competitors rank better than I do?
  • Which technical issues did the latest crawl find, and what should I fix first?
  • How often is my site cited or mentioned in ChatGPT answers and in Google AI Overviews?
  • Has my crawl or rank check finished?
  • Optional, uses credits: keyword ideas with search volume, difficulty and intent.

Tools

ToolPurposeCosts
list_projectsProjects of your account with IDs and domainsfree
get_project_overviewAudit score, rankings, AI visibility, tasks and next stepsfree
get_keyword_rankingsTracked keywords with positions and changesfree
get_search_performanceSearch Console or Bing clicks, impressions, CTR, positionfree
find_keyword_opportunitiesQueries on positions 4 to 20 with impressionsfree
analyze_competitorsVisibility and head-to-head comparison from stored resultsfree
get_crawl_issuesIssues of the latest crawl by severityfree
get_ai_visibilityCitations and mentions in ChatGPT answers and Google AI Overviewsfree
get_job_statusStatus and result of a background jobfree
get_seo_recommendationsPrioritised recommendations with impact, effort and evidencefree
get_codebase_overviewRoutes, metadata and live-page mapping from the codebase snapshotfree
get_conversion_performanceSessions, conversions and conversion rate per landing pagefree
research_keywordsKeyword ideas and metricscredits

All lists contain at most 25 rows. Tools only read data of projects you can access, and they never change your website.

ChatGPT does not offer research_keywords. In all other clients it needs an active Starter or Pro plan and the workspace setting “Agents can spend credits”. A daily credit limit for agents applies, and credits come from your monthly plan allowance first.

Interactive cards

In apps that can display MCP Apps, such as ChatGPT and Claude, four tools show their result as an interactive card: get_keyword_rankings, get_crawl_issues, find_keyword_opportunities and get_ai_visibility. Every other client, for example Claude Code, Cursor, Codex and the Gemini CLI, gets the same data as text. The cards load no external resources and open links only to the Serpel dashboard.

Set up

The server address is https://serpel.app/mcp (Streamable HTTP). Clients that support OAuth open a browser window for the sign-in on first use. Alternatively create an API token with the mcp:read permission (plus the read permissions for the data you want to use) in the Serpel dashboard under Settings, API tokens, and send it as Authorization: Bearer <token>.

Claude Code
claude mcp add --transport http serpel https://serpel.app/mcp

To load the plugin with its skill from a local checkout: claude --plugin-dir ./serpel-plugin.

ChatGPT and Codex

Install the plugin from the plugin directory, or add the remote MCP server https://serpel.app/mcp in the connector settings and sign in with your Serpel account.

Cursor

Add the server in .cursor/mcp.json or install the plugin from the marketplace:

{
  "mcpServers": {
    "serpel": { "url": "https://serpel.app/mcp" }
  }
}
VS Code with GitHub Copilot

Add the server in .vscode/mcp.json or install it from the MCP gallery (@mcp):

{
  "servers": {
    "serpel": { "type": "http", "url": "https://serpel.app/mcp" }
  }
}
Gemini CLI
gemini extensions install https://github.com/Serpelapp/serpel-plugin

Files

FileFor
plugin.json, mcp.json, skills/serpel/SKILL.mdAgent Plugins 1.0: ChatGPT, Codex, Cursor, GitHub Copilot, VS Code
chatGpt/skills/serpel/SKILL.mdSkill for the ChatGPT app directory, without keyword research, which ChatGPT does not offer
.claude-plugin/plugin.json, .mcp.jsonClaude Code
gemini-extension.jsonGemini CLI
server.jsonOfficial MCP registry, name app.serpel/mcp
assets/Icon

All manifests carry the same version.

Privacy and support

Tool results contain data from your Serpel account only. Disconnect a client at any time in the Serpel dashboard under Settings, API, connected apps.

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.