Skip to content
MCP server

MCP

MCP server for checking how ChatGPT and Google AI Mode name and cite brands

Status and history

Answered our latest tool-list check 22 hours ago with 6 tools. It will be marked verified after 3 successful checks in 7 days.

Install
Pricing
Freemium
Platforms
API, CLI, Web
Runs
Locally or Remotely
Version
1.1.0
Listing updated
3 days ago

About MCP

MCP is a freemium MCP server for marketing and seo. It can run locally on your machine or as a hosted remote server.

Parse collects answers from AI engines and the pages they cite, and this server lets an assistant query that data. A first call of check_brand works without signing in and shows the public view: figures over a 180-day window, at most 5 rows per list and no cursors. Signing in with Parse or sending a workspace key (parse_mcp_…) unlocks the full set of tools, starting with get_brief.

Tools cover searching and reading AI answers, the pages they cite and retrieved result lists, computing metrics such as share of answers naming a brand, and exporting long results to a file. Paid tools can ask AI engines live, fetch pages, apply a rubric, run a brand baseline or track topics and questions. Reading what Parse already holds is free; anything bought or fetched costs $0.01 per unit, and a call that would exceed the workspace budget returns a quote instead of buying.

The same tools are available over REST and through a one-file Node CLI. Replies include context, suggested next calls and paging, and values that are missing are marked as missing rather than zero.

Install

Run in a terminal

claude mcp add --transport http mcp https://agent.parse.gl/mcp

Add to Settings → Connectors → Add custom connector

https://agent.parse.gl/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "mcp": {
      "url": "https://agent.parse.gl/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "mcp": {
      "type": "http",
      "url": "https://agent.parse.gl/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "mcp": {
      "serverUrl": "https://agent.parse.gl/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "mcp": {
      "type": "http",
      "url": "https://agent.parse.gl/mcp"
    }
  }
}

Tools this server provides (6)

Read directly from the server.

check_brand
First question about one brand: how often AI mentions it and who is picked instead. Takes the brand's website (best), its name or a brd_ id, and answers in one call. For a brand Parse knows well: its share named and recommended in each of its markets, the brands named most in the market where it is strongest, and the topic it loses on. For one it knows little or nothing about: that fact, and the s
resolve_ids
kinds: brand, topic, question, market, page, site, place, or facet, surface, measure for the dictionary. Turn brand names, domains, URLs, topic, market or place names ("Austin, TX") or question wordings into Parse ids (brd_, top_, mkt_, plc_, qst_, pg_, site:), up to 25 at once: why each matched, its score, whether this workspace watches it (tracked), whether Parse has enough on a brand (indexed,
query_metrics
Count, share, trend: answers, share named / recommended / first, position, tone and treatment mix, citations, pages and sites cited, comparisons won and lost (one brand: subject), search_reach (a brand's sites in retrieved result lists; no by), memory_gap (a brand named with search on against search off, on the questions both answered; no by) and concentration (answers naming the top 1, 3, 5 brand
search_answers
Search AI answers by brand (named, recommended, first, not named, warned), topic, market, question, engine, place, country asked from, site or page cited, keywords (filter.text) or any facet: compact rows with ans_ ids. view: "diff" shows which brands and cited pages came and went for up to five questions or one topic. For counts and shares use query_metrics. Free: with credits every row; without,
search_cited_pages
Pages AI answers cite, by brand, topic, market, question, engine, site or facet. For outreach pass pages = third_party: most cited pages are brands' own sites, and this leaves them out. Each row: citations, answers and questions citing it (over the whole filter), weeks cited, answersCiting, answersNaming and answersOmitting (over every answer citing it, the brand rule aside: not the same answers a
get_brief
Needs a free Parse sign-in, which this call asks the person for: the person's own brand's brief (what changed, what to do next) and their work record. Call it only when the person wants their own brand, tracking or their record, or says yes to an offer; for any public question use check_brand and the other tools.

What it does

Public brand check
check_brand works logged out and returns a limited public view.
Answer and citation search
Searches AI answers, cited pages and retrieved result lists.
Metrics over AI answers
query_metrics computes shares with 95% ranges and matched-question changes.
Budgeted paid tools
Live checks, page fetches and rubric judgments run within a workspace budget, with quotes first.
Tracking and briefs
Tracks brands, topics and questions and summarises changes with get_brief.
REST and CLI
Same tools at agent.parse.gl/v1/tools and via a single-file Node CLI.

Good for

  • Check whether an AI engine names a brand when asked about its category
  • Find which pages AI answers cite for a set of questions
  • Track a brand's AI answer share weekly and log what changed
  • Export a large set of AI answers for analysis

Pricing

Freemium

Reading collected data is free; paid checks and fetches cost $0.01 each, first 500 checks free

Prices change. Check parse.gl before you buy.

MCP FAQ

Do I need an account?

Not for the public view: check_brand and a few other tools work logged out with limited rows. The rest need a Parse sign-in or a workspace key.

What does it cost?

Reading existing data is free; checks, page fetches and judgments cost $0.01 each, and the first 500 checks are free with no card per the pricing page.

How do I connect?

Run claude mcp add --transport http parse-agent https://agent.parse.gl/mcp, optionally with an Authorization Bearer key.

What are the rate limits?

A workspace gets 600 calls a minute; logged out is 40 calls a session and 300 a day per address.

Is MCP free?

MCP has a free plan, with paid plans for more usage or features.

How do I add MCP to Claude Code?

Run this command in your terminal: claude mcp add --transport http mcp https://agent.parse.gl/mcp

What are the best alternatives to MCP?

The closest alternatives to MCP are Agent-Native Analytics, Google Ads + Meta Ads + SEO/GEO MCP, Tapetide - Indian Stock Market MCP, World Monitor and Calven.

Alternatives to MCP

  1. Open-source analytics app where an AI agent builds charts and dashboards from your data

    • Remote
    • Open source
    • 7.7K GitHub stars
    • Analytics and data
  2. Hosted MCP connector for managing Google Ads, Meta Ads, GA4 and Search Console from chat

    • 8 tools
    • Remote
    • Freemium
    • 8.1K GitHub stars
    • Marketing and SEO
  3. Indian NSE and BSE stock market data for AI assistants through 48 MCP tools

    • 17 tools
    • Remote
    • Freemium
    • 150 GitHub stars
    • Finance and payments
  4. World Monitor

    MCP server

    Live geopolitical intelligence dashboard with an API and MCP server for agents

    • Remote
    • Freemium
    • Analytics and data
  5. Calven

    MCP server

    Product marketing context from Calven delivered into Claude, ChatGPT and other AI tools

    Not checked yet
    • Remote
    • Free trial
    • 3 GitHub stars
    • Marketing and SEO
  6. CompanyData

    MCP server

    Search and enrich company records from official registers via the CompanyData API

    • 3 tools
    • Remote
    • Paid
    • 3 GitHub stars
    • Analytics and data