This page is loading and it needs JavaScript. The written page below is served without it.
Parlay Builder publishes an MCP server: a small, read-only endpoint that an AI client can query directly. Add it once to Claude or ChatGPT and the assistant can pull tonight’s ranked prop board, how often a player has actually cleared a line, and the graded record of what the model has published, without you leaving the chat to look any of it up.
MCP is the Model Context Protocol, an open standard Anthropic published in 2024 for connecting AI assistants to outside data. A “connector” or “MCP server” is one such source. The client asks it questions in a fixed shape and it answers with structured data instead of a web page, which is what lets an assistant reason over the numbers rather than scrape them out of prose.
Every tool the connector serves is listed below, with the inputs it takes and the limits that apply to what it returns. The list is generated from the server’s own tool definitions, so it cannot describe a tool or an input the server does not have. Every tool is a read, and none of them returns a sportsbook price.
pb_top_props
“What’s on the NBA board tonight?”
The ranked prop board for a league and date, filtered the way you would filter it on the site: league, market, minimum sample, how far out the game is. Each row carries its hit rate, its sample and its historical band.
league League, required: one of NBA, NFL, MLB or WNBA.date Slate date, optional: a date, written YYYY-MM-DD.days_ahead Extra days past the date, optional: a whole number from 0 to 7.market Market, optional: text.min_games Fewest games behind a rate, optional: a whole number from 0 to 500.min_hit_rate Lowest hit rate, in percent, optional: a number from 0 to 100.limit How many rows, optional: a whole number from 1 to 100.offset Rows to skip, for paging, optional: a whole number, 0 or more.pb_player_hit_rate
“How often has Judge cleared 0.5 hits?”
One player, one market, one threshold, measured across five windows (last 3, last 5, last 10, this season and last season) each with the games behind it, plus trend, consistency and current streak.
league League, required: one of NBA, NFL, MLB or WNBA.player Player name, required: text.market Market, required: text.line Line, required: a number.side Side, optional: one of over or under.pb_compare_props
“Which of these three has cleared it most?”
Two to six props measured side by side on one window, so the comparison is like for like rather than three separate answers a reader has to line up.
league League, required: one of NBA, NFL, MLB or WNBA.props Props to compare, required: a list of 2 to 6 entries, each with player, market, line and side.window Window to compare on, optional: one of last3, last5, last10, current_season or last_season.pb_most_consistent
“Who is steadiest on the MLB board?”
Board rows ordered by the measured consistency index (how little a player’s output has moved around its own average) on the same thresholds the chat uses.
league League, required: one of NBA, NFL, MLB or WNBA.date Slate date, optional: a date, written YYYY-MM-DD.days_ahead Extra days past the date, optional: a whole number from 0 to 7.market Market, optional: text.limit How many rows, optional: a whole number from 1 to 50.pb_track_record
“Does any of this actually work?”
The forward record: every prop the model has published since inception, graded against the final box score, broken out by league, market and confidence band.
league League, optional: one of NBA, NFL, MLB or WNBA.since Earliest slate date, optional: a date, written YYYY-MM-DD.pb_board_coverage
“How much of tonight is covered?”
How many of the slate’s games and props the board actually has measured history for, which is the number that tells you whether an empty answer means a quiet night or a thin feed.
league League, optional: one of NBA, NFL, MLB or WNBA.hours_ahead Hours to look ahead, optional: a whole number from 1 to 168.pb_explain_metric
“What does this number mean?”
A plain definition of any figure the other tools return: what it counts, over what window, and what it does not claim.
metric Which figure to define, optional: one of hit_rate, sample_size, settles_at, consistency_index, trend_direction, streak, price_status, historical_grade, board_rank or track_record.pb_team_market
“How often have the Bulls covered 3.5?”
One team, one market (won, covered a spread, or its own points over or under a line), across five windows, overall or home or away, each with the games behind it.
league League, required: one of NBA, NFL, MLB or WNBA.team Team name, required: text.market Market, required: one of moneyline, spread or team_total.line Line, optional: a number.side Side, optional: one of over or under.location Home, away or overall, optional: one of overall, home or away.pb_whos_playing
“Who plays tonight, and who is on the board?”
The scheduled games for a league and date, and for each game the players with a prop on the board. Not a lineup and not an injury report: there is no injury feed, and it says so.
league League, required: one of NBA, NFL, MLB or WNBA.date Slate date, optional: a date, written YYYY-MM-DD.days_ahead Extra days past the date, optional: a whole number from 0 to 7.team Team name, optional: text.pb_explain_grade
“Why is this prop a C?”
Both the rules and the letter. For one prop on the board: its historical band, a letter from A (cleared 80% of the time or more) to F (under 50%), every rule that held it lower in plain words, and the rate and games it was read from. A prop not on the board gets the band thresholds and no letter. A description of the past, not a forecast.
league League, required: one of NBA, NFL, MLB or WNBA.player Player name, required: text.market Market, required: text.line Line, required: a number.side Side, optional: one of over or under.date Slate date, optional: a date, written YYYY-MM-DD.days_ahead Extra days past the date, optional: a whole number from 0 to 7.pb_research_team
“What does the Celtics’ profile look like?”
A team’s pace and scoring against the league average, its defense against each position by stat with the league rank, and its win record across windows.
league League, required: one of NBA, NFL, MLB or WNBA.team Team name, required: text.The connector is included with the subscription, $7.99 a month or $79.99 a year, and the plan check runs on the server every time a key is used rather than only when one is issued.
On the setup page, once signed in. One key per account, shown once at the moment it is created: we store only a hash of it and cannot show it to you again. Lose it and you revoke and regenerate, which is a thirty-second job.
The endpoint with your key already appended. That is the whole of the setup and the whole of the authentication: no OAuth fields to fill in, no config file to edit unless you want one. Treat the pasted URL like a password, because it carries one.
A small endpoint that answers an AI client’s questions with structured data instead of a web page, using the open Model Context Protocol, so the assistant reasons over real fields rather than over whatever it managed to scrape.
No, and that is deliberate rather than a gap. Every posted price, book name and price age is stripped from the response before it leaves, because our data provider permits displaying prices in a product and prohibits re-serving them as raw data. The connector tells your assistant whether a book has the selection priced and links to the page here where the number is shown.
No. No response carries a pick rating, an edge figure or a projected return: those fields are removed inside the response builder, so a leak is an error rather than a quiet regression. The one letter it sends is the historical band, which describes how often a line has been cleared in the past and never tonight. What comes back is how often something has happened, over how many games, and where the board ranked it. Deciding what that is worth is still yours.
Claude, which it was built and verified against, and ChatGPT behind Developer mode. The consumer Gemini app has no way to add a custom MCP server, so it cannot; Gemini Enterprise can, added by an administrator as a custom MCP server data store.
Every tool is a read. The connector cannot place a bet, move money, see a payment method or change anything in your account, and the key is scoped to you: it reaches no other subscriber’s data. The key does travel in the URL you paste, because no hosted AI client can send a custom header, so treat that URL like a password and revoke it if it ever ends up somewhere it should not.
Yes. The connector is included with the subscription, $7.99 a month or $79.99 a year, and the plan check runs on the server on every call rather than only when a key is issued. Without a subscription you can still open the research pages and the top of the board.
You revoke it and generate another on the setup page. We store only a hash of a key and genuinely cannot show it to you a second time, so there is no “reveal” that would recover it. Revoking invalidates whatever client was using it, which is the point.
Odds and lines come from licensed data feeds and refresh on a schedule; hit rates and grades are Parlay Builder’s own calculations.