PuckAPI

This connector has been deprecated

Moved to a server instead of a connector

get_odds

Read-only

Use when checking betting lines. Pass game_id for one game (10 credits): moneyline, puck line and total by bookmaker, opening and closing. Or pass season (optionally with team) for a whole season of closing lines in one call, with final scores, at 10 credits per game returned (a full season is ~13,700 credits, a team-season ~820), 100 games a page by default (up to 500). Each book's line is an array in the order given by columns. Set estimate_only to get the price without being charged. Every game from 2019-20 on (and part of the 2019 playoffs): about 12-14 books a game from 2020-21 to 2023-24, 6-7 in 2019-20, 4 in 2024-25, 60+ in 2025-26 with all_books. A season those four do not carry (2019-20) returns every book it has. Each line says its source: the_odds_api (our Odds API pulls, from August 2020) or espn (ESPN's per-book lines); a season answer lists each game's sources.

Input Schema

TableJSON Schema

Name	Required	Description
`team`	No	With season: only this team's games (abbreviation, e.g. BUF).
`limit`	No	With season: games per page (default 100, max 500).
`offset`	No	With season: page offset; use next_offset from the previous page.
`season`	No	Season for a whole season of lines, e.g. 2025-26 or 20252026. Used when game_id is not set.
`game_id`	No	NHL game ID (e.g. 2025020887). Omit and pass season for a whole season.
`all_books`	No	With season: every book held (12-14 a game from 2020-21 to 2023-24, more than 60 on about 300 games in 2025-26) instead of DraftKings, FanDuel, BetMGM and ESPN BET.
`bookmaker`	No	Filter by bookmaker key (e.g. draftkings, fanduel, betmgm). Lowercase.
`estimate_only`	No	Return what this call would cost, without charging or running it.
`snapshot_type`	No	opening (game-day morning; for a game not yet played, the latest morning line) or closing (final pre-game). Single game: default both. Season: default closing.

Output Schema

TableJSON Schema

Name	Required	Description	Default
`note`	No
`odds`	No
`tool`	No
`count`	No
`games`	No
`season`	No
`columns`	No
`credits`	No
`game_id`	No
`away_team`	No
`game_date`	No
`home_team`	No
`next_offset`	No
`total_games`	No
`estimate_only`	No
`snapshot_type`	No
`bookmaker_count`	No
`credits_per_game`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 8, 2026

changedInput schema / properties / all_books / description

Previous value: -"With season: every book held (12-14 a game from 2020-21 to 2023-24, 60+ from 2025-26) instead of DraftKings, FanDuel, BetMGM and ESPN BET."New value: +"With season: every book held (12-14 a game from 2020-21 to 2023-24, more than 60 on about 300 games in 2025-26) instead of DraftKings, FanDuel, BetMGM and ESPN BET."

Changed1 schema field changedOct 8, 2026
- addedOutput schema / properties / note
```
Added value: +{
+  "type": "string"
+}
```

Changed1 schema field changedOct 8, 2026

changedInput schema / properties / all_books / description

Previous value: -"With season: every book held (60+ from 2025-26) instead of DraftKings, FanDuel, BetMGM and ESPN BET."New value: +"With season: every book held (12-14 a game from 2020-21 to 2023-24, 60+ from 2025-26) instead of DraftKings, FanDuel, BetMGM and ESPN BET."

Changed1 schema field changedSep 28, 2026

changedInput schema / properties / snapshot_type / description

Previous value: -"opening (game-day morning) or closing (final pre-game). Single game: default both. Season: default closing."New value: +"opening (game-day morning; for a game not yet played, the latest morning line) or closing (final pre-game). Single game: default both. Season: default closing."

Changed21 schema fields changedSep 27, 2026

addedInput schema / properties / all_books

Added value: +{
+  "description": "With season: every book held (60+ from 2025-26) instead of DraftKings, FanDuel, BetMGM and ESPN BET.",
+  "type": "boolean"
+}

addedInput schema / properties / estimate_only

Added value: +{
+  "description": "Return what this call would cost, without charging or running it.",
+  "type": "boolean"
+}

changedInput schema / properties / game_id / description

Previous value: -"NHL game ID (e.g. 2025020887)"New value: +"NHL game ID (e.g. 2025020887). Omit and pass season for a whole season."

addedInput schema / properties / limit

Added value: +{
+  "description": "With season: games per page (default 100, max 500).",
+  "maximum": 500,
+  "minimum": 1,
+  "type": "integer"
+}

addedInput schema / properties / offset

Added value: +{
+  "description": "With season: page offset; use next_offset from the previous page.",
+  "maximum": 9007199254740991,
+  "minimum": 0,
+  "type": "integer"
+}

addedInput schema / properties / season

Added value: +{
+  "description": "Season for a whole season of lines, e.g. 2025-26 or 20252026. Used when game_id is not set.",
+  "type": "string"
+}

changedInput schema / properties / snapshot_type / description

Previous value: -"Filter by snapshot type: opening (game-day morning lines) or closing (final pre-game lines). Default returns both."New value: +"opening (game-day morning) or closing (final pre-game). Single game: default both. Season: default closing."

addedInput schema / properties / team

Added value: +{
+  "description": "With season: only this team's games (abbreviation, e.g. BUF).",
+  "type": "string"
+}

removedInput schema / required
```
Removed value: -[
-  "game_id"
-]
```

addedOutput schema / properties / columns

Added value: +{
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}

addedOutput schema / properties / count
```
Added value: +{
+  "type": "number"
+}
```
addedOutput schema / properties / credits
```
Added value: +{
+  "type": "number"
+}
```
addedOutput schema / properties / credits_per_game
```
Added value: +{
+  "type": "number"
+}
```
addedOutput schema / properties / estimate_only
```
Added value: +{
+  "type": "boolean"
+}
```

addedOutput schema / properties / games

Added value: +{
+  "items": {
+    "additionalProperties": {},
+    "propertyNames": {
+      "type": "string"
+    },
+    "type": "object"
+  },
+  "type": "array"
+}

addedOutput schema / properties / next_offset

Added value: +{
+  "type": [
+    "number",
+    "null"
+  ]
+}

addedOutput schema / properties / season
```
Added value: +{
+  "type": "string"
+}
```
addedOutput schema / properties / snapshot_type
```
Added value: +{
+  "type": "string"
+}
```
addedOutput schema / properties / tool
```
Added value: +{
+  "type": "string"
+}
```
addedOutput schema / properties / total_games
```
Added value: +{
+  "type": "number"
+}
```

removedOutput schema / required

Removed value: -[
-  "game_id",
-  "home_team",
-  "away_team",
-  "game_date",
-  "odds",
-  "bookmaker_count"
-]

First observedSep 27, 2026

TDQS

A4.1/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare readOnlyHint=true and openWorldHint=false; the description adds substantial non-obvious behavior: credit pricing (10/game, ~13,700 for a season), that estimate_only avoids charging, per-season book coverage counts, and provenance tagging (the_odds_api vs espn). This is exactly the cost/quota/disclosure context annotations cannot carry.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded correctly with when-to-use and cost, but the middle is a dense run-on of parenthetical figures (cost totals, per-season book counts) and the book-count figures are duplicated verbatim in the all_books schema description. The provenance sentence at the end is also clipped and hard to parse.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 9-parameter tool with two operating modes and an output schema already covering return values, the description supplies mode routing, pricing, pagination defaults, and data-coverage caveats. Only the error behavior when neither game_id nor season is supplied is left unstated.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds cross-parameter semantics the schema only hints at: game_id and season are alternatives ('used when game_id is not set'), team/all_books/snapshot_type only apply in season mode, and each book's line array follows the columns order.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a clear verb+resource (retrieve betting lines/odds) and goes further by spelling out the two retrieval modes and what each returns. It never distinguishes itself from the sibling get_line_movement, which an agent could easily confuse with odds retrieval, so it stops short of a 5.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit mode-selection guidance ('Pass game_id for one game ... Or pass season ... for a whole season'), plus the cost-saving estimate_only path. It does not name or exclude any sibling (e.g. get_line_movement for movement over time), so usage is clear but has no explicit alternatives.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All PuckAPI tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?