Skip to main content
Glama

mcp-sales-intel

A production-shaped MCP server that turns any marketplace into competitive intelligence.

Scout live Fiverr gigs, analyse pricing bands and competition, and get a data-anchored package-price ladder — all through natural language from Claude, Cursor, or any MCP client.

This is the reference implementation I use to demonstrate MCP server work: real external data, bounded inputs, structured errors, caching, and both stdio and streamable-HTTP transports. Zero secrets required.


Why this exists

Most Fiverr MCP gigs ship a demo server that returns hardcoded JSON. This one actually scrapes, actually parses, actually computes. It's the difference between "here's a server" and "here's a server that works on Monday."

Related MCP server: fiverr-mcp

Tools

Tool

What it does

list_gigs

Scrape live Fiverr search results → title, seller, rating, reviews, entry price, URL. Filter by price/rating.

analyse_market

Pricing bands, review distribution, and an explicit "open market" verdict from raw page markdown.

price_gig

A Basic/Standard/Premium ladder anchored to observed medians — not vibes.

Quick start

# stdio (works with Claude Desktop, Cursor, Hermes, mcporter)
uv run mcp-sales-intel

# streamable HTTP (for shared/remote deployments)
MCP_TRANSPORT=streamable-http uv run mcp-sales-intel

Wire it into an MCP client

claude_desktop_config.json / Cursor mcp.json:

{
  "mcpServers": {
    "sales-intel": {
      "command": "uv",
      "args": ["run", "--directory", "/abs/path/to/mcp-sales-intel", "mcp-sales-intel"]
    }
  }
}

Requirements

  • Python 3.12+

  • mcp Python SDK 2.x (note: FastMCP was renamed MCPServer in 2.0)

  • firecrawl CLI on PATH — only for live scraping. The other two tools work without it.

Example

"Analyse the mcp server development market and price my gig."

The agent calls list_gigs, feeds the result to analyse_market, then price_gig, and returns a real ladder with the reasoning attached.

Design notes

  • Bounded inputs. Query length and max_results are clamped; bad ranges return a structured error instead of raising.

  • 30-minute cache per query, LRU-evicted, so repeated agent calls don't re-scrape and burn credits.

  • Honest failures. Missing CLI, timeouts, non-zero exits, and layout drift all return {"error": ..., "gigs": []} rather than fabricating results.

License

MIT

Available Tools

3 tools
analyse_marketAnalyse a market from scraped markdownA

Take raw Fiverr search-page markdown (as a string) and compute pricing bands, review-count distribution and an 'open market' verdict — i.e. how reachable a new seller is. Use for competitive analysis without re-scraping.

Inputs: markdown (required), open_market_review_threshold. Returns: object

list_gigsList gigs in a categoryB

Scrape live Fiverr search results for a query and return gig cards with title, seller, rating, review count, entry price and URL. Use this to scout what sellers are actually offering and charging.

Inputs: query (required), max_price, min_price, min_rating, +1 more. Returns: object

price_gigRecommend a gig price ladderB

Given a set of gigs (or a market query), recommend a three-tier Basic/Standard/Premium package ladder with prices anchored to observed medians.

Inputs: gigs, query, floor_usd. Returns: object

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 3 tools

    Disambiguation5/5

    The three tools target clearly distinct actions: scraping live gigs, analyzing supplied market markdown, and generating a pricing ladder. Their descriptions make boundaries explicit, so an agent should not confuse list_gigs with analyse_market or price_gig.

    Naming Consistency5/5

    All names follow a consistent snake_case verb_noun pattern: list_gigs, analyse_market, price_gig. The convention is predictable and readable throughout.

    Tool Count4/5

    Three tools cover the core scrape-analyze-price workflow without redundancy, which is well-scoped for a niche sales-intel server. It is slightly lean, but each tool earns its place.

    Completeness3/5

    The surface covers scraping, analysis, and pricing, but there is a notable gap: analyse_market requires raw Fiverr search-page markdown while list_gigs only returns structured gig cards, leaving no direct tool path to obtain that input. Additional gaps include seller-level detail and historical tracking, though the main workflow is partially covered.

    Maintenance

    ActivitySlowing
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      C
      maintenance
      MCP server for searching Fiverr gigs, comparing freelancer packages, reading seller reviews, and exploring service categories. Provides structured tools for gig search, details, seller profiles, reviews, and category browsing with built-in anti-bot handling.
      5
      18
      MIT
    • F
      license
      Not graded
      quality
      D
      maintenance
      Enables AI assistants to perform real-time market research including competitor analysis, pricing intelligence, and company overviews via live web search through Tavily API.
      -
    • F
      license
      B
      quality
      C
      maintenance
      Enables AI-powered competitor pricing analysis by scraping LLM inference pricing websites via Firecrawl, storing the data in SQLite, and answering cost comparison queries through natural language.
      2
      -