Sports Card Agent

io.github.rjexilePublished by GitHub account rjexilev1.0.1Updated Oct 24, 2026

Sports card pricing, market analysis, grading ROI, and player stats (NBA/NFL/MLB)

VerifiedSTDIODesktop onlyFinanceData & Analytics

Overview

AI-generated overview

Lets an assistant look up sports trading card prices, market trends, grading ROI, and NBA/NFL/MLB player stats.

What you can do
Provides nine tools for sports card research: eBay sold and active price lookups, market trend analysis with arbitrage detection, grading ROI calculations for PSA, BGS and SGC, buy/sell/hold investment advice, trending player alerts, and vintage card analysis across 40+ pre-2000 sets. Player stat tools cover NBA, NFL and MLB performance with card market context. It runs locally and can start with mock data when no API keys are set.
When to use it
Useful when an assistant is asked to price a specific card, judge whether grading is worth the fee, compare raw versus graded values, or connect a player's recent performance to card market movement. Suited to collectors and resellers doing repeated card research rather than one-off lookups.
Requirements
Runs as a local Python process installed from PyPI (pip install sports-card-agent) over stdio, on desktop only. Optional credentials: BALLDONTLIE_API_KEY for player stats, and EBAY_APP_ID plus EBAY_CERT_ID for eBay pricing. Without keys it falls back to mock data. Network access is needed for live data.
Before you install
Three secrets are requested: BALLDONTLIE_API_KEY, EBAY_APP_ID and EBAY_CERT_ID; store them in the environment or a .env file and never share them. Without keys the server returns mock data, so prices and recommendations may be fabricated rather than real. Investment advice and arbitrage signals are generated output, not verified financial guidance, and card pricing depends on third-party eBay and stats APIs.

Installation

In SourceWeft

  1. Open Sports Card Agent in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

Sports Card Agent

An MCP server that gives AI agents expert-level sports trading card data. Covers pricing, market analysis, arbitrage detection, grading ROI, investment advice, player stats (NBA/NFL/MLB), vintage card analysis, and trending player alerts.

9 tools. 3 sports. 40+ vintage sets. Zero manual research.

Tools

Pricing & Market

ToolDescription
card_price_lookupReal-time sold and active prices from eBay. Supports any sport, brand, year, or grading.
card_market_analysisTrend analysis comparing sold vs asking prices. Detects arbitrage opportunities where cards are listed below market value.

Player Stats

ToolDescription
player_stats_lookupMulti-sport player stats (NBA/NFL/MLB) with card market insights based on performance.
nfl_stats_lookupNFL passing, rushing, receiving, and defensive stats with card market insights.
mlb_stats_lookupMLB batting (AVG, HR, RBI, OPS) and pitching (ERA, K, WHIP) stats with card insights.

Analysis & Strategy

ToolDescription
grading_roi_calculatorCalculates whether grading a card is profitable. Compares raw vs graded prices for PSA, BGS, and SGC with fee-adjusted ROI.
card_investment_advisorBuy/sell/hold recommendations combining market trends with player performance data across all 3 sports.
trending_playersIdentifies NBA players with breakout performances whose cards are likely rising in value.
vintage_card_analysisEra-specific analysis for pre-2000 cards. Covers 40+ iconic sets from 1909 T206 to 2000 Playoff Contenders with grade-based pricing.

Quick Start

Install from PyPI

bash
pip install sports-card-agent

Run the server

bash
sports-card-agent

Use with Claude Desktop

Add to your claude_desktop_config.json:

json
{  "mcpServers": {    "sports-card-agent": {      "command": "sports-card-agent"    }  }}

Use with Claude Code

Add to your .mcp.json:

json
{  "mcpServers": {    "sports-card-agent": {      "command": "sports-card-agent"    }  }}

Configuration

Create a .env file or set environment variables:

bash
# eBay API (register free at developer.ebay.com)EBAY_APP_ID=your_app_idEBAY_CERT_ID=your_cert_id
# Ball Don't Lie API (register free at app.balldontlie.io)BALLDONTLIE_API_KEY=your_api_key

The server works without API keys using mock data, so you can try it immediately.

Example Queries

Once connected, any AI agent can ask:

  • "What's a 2023 Topps Chrome Wembanyama rookie selling for?"
  • "Should I buy or sell my Patrick Mahomes rookie card?"
  • "Is it worth grading my 1986 Fleer Jordan?"
  • "Who are the trending NBA players whose cards are rising?"
  • "Analyze the market for Ken Griffey Jr 1989 Upper Deck rookie"
  • "What's the investment outlook on vintage 1952 Topps Mickey Mantle?"
  • "How is Shohei Ohtani performing this season and what does that mean for his cards?"

Sports & Sets Covered

Sports: Baseball, Basketball, Football, Hockey, Soccer

Player Stats: NBA (all teams), NFL (all positions), MLB (batting + pitching)

Vintage Sets Include: 1909 T206, 1933 Goudey, 1951 Bowman, 1952 Topps, 1954-55 Topps, 1958 Topps Football, 1961 Fleer Basketball, 1965 Topps Football, 1966 Topps Hockey, 1969 Topps, 1979 O-Pee-Chee, 1981 Topps Football, 1984 Topps Football, 1986 Fleer Basketball, 1986 Donruss, 1989 Upper Deck, 1993 SP, 1996 Topps Chrome, 1997 Metal Universe, 2000 Playoff Contenders, and more.

Grading Companies: PSA, BGS, SGC (all service tiers with current pricing)

Development

bash
git clone https://github.com/rjexile/sports-card-agent.gitcd sports-card-agentpython -m venv venvsource venv/Scripts/activate  # Windowspip install -e .python test_all.py  # Run all 29 tests

License

MIT

Source: README.md at commit 9e59a60

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v1.0.1LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts