
Company News
Jerod Santo Joins Socket as Head of Media
Allow myself to introduce... myself.
@davincibets/mcp-server
Advanced tools
MCP server for Da Vinci Bets — AI sports predictions in your AI assistant
Give your AI assistant native access to AI-powered sports predictions.
Works with Claude Desktop, Claude Code, Cursor, Windsurf, and any MCP-compatible client.
Go to davincibets.io/settings → API tab → Generate Key.
Your key looks like: dvb_live_a3f2... — copy it immediately (shown once).
Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or ~/.config/Claude/claude_desktop_config.json (Linux) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"davinci-bets": {
"command": "npx",
"args": ["-y", "@davincibets/mcp-server@latest"],
"env": {
"DVB_API_KEY": "dvb_live_your_key_here"
}
}
}
}
claude mcp add davinci-bets -- npx -y @davincibets/mcp-server@latest
Then set the environment variable:
export DVB_API_KEY="dvb_live_your_key_here"
Or add to your project's .mcp.json:
{
"mcpServers": {
"davinci-bets": {
"command": "npx",
"args": ["-y", "@davincibets/mcp-server@latest"],
"env": {
"DVB_API_KEY": "dvb_live_your_key_here"
}
}
}
}
Open Settings → MCP Servers → Add:
davinci-betsnpx-y @davincibets/mcp-server@latestDVB_API_KEY=dvb_live_your_key_hereAdd to your Windsurf MCP config:
{
"mcpServers": {
"davinci-bets": {
"command": "npx",
"args": ["-y", "@davincibets/mcp-server@latest"],
"env": {
"DVB_API_KEY": "dvb_live_your_key_here"
}
}
}
}
Once connected, you can ask naturally:
get_predictionsFetch AI predictions by sport, market, and confidence.
| Parameter | Type | Default | Description |
|---|---|---|---|
sport | NBA, NFL, NHL, MLB | all | Filter by sport |
market | moneyline, spread, totals | moneyline | Betting market |
team | string | — | Filter by team name, abbreviation, or city (e.g., Lakers, LAL, Boston) |
sort | confidence, edge, game_time | confidence | Sort order |
limit | 1-50 | 25 | Max results |
min_confidence | 0-100 | 60 | Minimum confidence % |
get_gameFind a specific game's prediction by team name. Intelligent lookup — give it a team name (nickname, abbreviation, or city) and it finds the matching game.
| Parameter | Type | Required | Description |
|---|---|---|---|
team | string | Yes | Team name, abbreviation, or city (e.g., Lakers, LAL, Los Angeles) |
opponent | string | No | Narrow to a specific matchup (e.g., Celtics) |
sport | NBA, NFL, NHL, MLB | No | Sport hint — auto-detected from team name if omitted |
market | moneyline, spread, totals | No | Market to look up (default: moneyline) |
lookup_teamsSearch the team registry by name, abbreviation, city, or sport. Instant local lookup — no API call, no rate limit impact.
| Parameter | Type | Required | Description |
|---|---|---|---|
query | string | No | Search by name, abbreviation, or city (e.g., NY, CHI, Boston) |
sport | NBA, NFL, NHL, MLB | No | Filter to a specific sport |
Covers 124 teams across 4 sports. Omit both parameters for a summary count.
get_top_picksGet the highest-conviction picks across all sports and markets.
| Parameter | Type | Default | Description |
|---|---|---|---|
count | 1-20 | 10 | Number of top picks |
min_confidence | 0-100 | 65 | Minimum confidence % |
get_scheduleGet today's game schedule with predictions for a sport.
| Parameter | Type | Required | Description |
|---|---|---|---|
sport | NBA, NFL, NHL, MLB | Yes | Which sport |
get_pick_cardRender a prediction as a shareable 1080x1350 PNG image. Returns a permanent public URL plus an inline image preview.
| Parameter | Type | Required | Description |
|---|---|---|---|
game_id | string | No | Game ID from predictions |
team | string | No | Alternative — pass a team name and the card for their next game is rendered |
sport | NBA, NFL, NHL, MLB | No | Sport hint |
Provide either game_id or team. The team option requires an extra API call to resolve.
get_batch_cardsRender multiple pick cards at once (up to 10). Designed for marketing teams.
| Parameter | Type | Required | Description |
|---|---|---|---|
game_ids | string[] | Yes | Array of game IDs (max 10) |
sport | NBA, NFL, NHL, MLB | No | Sport hint |
get_performanceGet current prediction performance snapshot.
| Parameter | Type | Default | Description |
|---|---|---|---|
sport | NBA, NFL, NHL, MLB | all | Filter by sport |
market | moneyline, spread, totals | all | Filter by market |
get_track_recordGet verified win rates, ROI, and historical performance from settled predictions. Shows actual results by sport, market, confidence tier, current streak, and recent settled picks. Public endpoint — does not count against your rate limit.
| Parameter | Type | Default | Description |
|---|---|---|---|
sport | NBA, NFL, NHL, MLB | all | Filter by sport |
days | 7-365 | 30 | Lookback period in days |
| Tier | Per Hour | Per Day | Who |
|---|---|---|---|
| Personal | 100 | 500 | $14.99/mo subscribers |
| Partner | 500 | 10,000 | Marketing agencies (DVB-Social) |
| Internal | Unlimited | Unlimited | Da Vinci Bets systems |
Rate limit info is included in prediction responses. When limits are hit, the server returns a clear error with retry timing.
DVB-Social and marketing partners get elevated rate limits for batch content generation:
# Example workflow in Claude:
1. "Get today's NBA moneyline predictions"
2. "Generate pick cards for all games above 70% confidence"
3. "Also get the NHL schedule and render those cards too"
The get_batch_cards tool can render up to 10 cards per call — ideal for daily social media pipelines.
Contact support@davincibets.io for partner API key provisioning.
| Variable | Required | Description |
|---|---|---|
DVB_API_KEY | Yes | Your API key (dvb_live_xxx) |
DVB_API_URL | No | Override API base URL (default: https://www.davincibets.io) |
# Clone and install
cd packages/mcp-server
npm install
# Run locally
DVB_API_KEY=dvb_live_xxx npm run dev
# Build
npm run build
MIT
FAQs
MCP server for Da Vinci Bets — AI sports predictions in your AI assistant
The npm package @davincibets/mcp-server receives a total of 88 weekly downloads. As such, @davincibets/mcp-server popularity was classified as not popular.
We found that @davincibets/mcp-server demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.

Company News
Allow myself to introduce... myself.

Research
/Security News
A Twitch browser extension on Chrome and Firefox forwards users’ live OAuth session tokens through proxies controlled by a Russian bot service.

Security News
Anthropic found biased reasoning and recklessness drove Claude Mythos 5 to publish malware on PyPI and compromise a security vendor.