Agent Skills

nba-data

researchmachina-sports

NBA data via ESPN public endpoints, the NBA live CDN, and NBA Stats (stats.nba.com) — scores, standings, rosters, schedules, game summaries, play-by-play, injuries, futures, depth charts, leaders, and news, plus an analytics backend: advanced ratings, per-shot court coordinates, career splits, and history to 1946. Zero config, no API keys. Use when: user asks about NBA scores, standings, team rosters, schedules, game stats, box scores, play-by-play, injuries, transactions, betting futures, depth

Install

npx skills add https://github.com/machina-sports/sports-skills --skill nba-data
SKILL.md

NBA Data

Before writing queries, consult references/api-reference.md for endpoints, ID conventions, and data shapes.

Setup

Before first use, check if the CLI is available:

which sports-skills || pip install sports-skills

If pip install fails with a Python version error, the package requires Python 3.10+. Find a compatible Python:

python3 --version  # check version
# If < 3.10, try: python3.12 -m pip install sports-skills
# On macOS with Homebrew: /opt/homebrew/bin/python3.12 -m pip install sports-skills

No API keys required.

Quick Start

Prefer the CLI — it avoids Python import path issues:

sports-skills nba get_scoreboard
sports-skills nba get_standings --season=2025
sports-skills nba get_teams

CRITICAL: Before Any Query

CRITICAL: Before calling any data endpoint, verify:

  • Season year is derived from the system prompt's currentDate — never hardcoded.
  • If only a team name is provided, call get_teams to resolve the team ID before using team-specific commands.

Choosing the Season

Derive the current year from the system prompt's date (e.g., currentDate: 2026-02-18 → current year is 2026).

  • If the user specifies a season, use it as-is.
  • If the user says "current", "this season", or doesn't specify: The NBA season runs October–June. If the current month is October–December, the active season year matches the current year. If January–June, the active season started the previous calendar year (use that year as the season).

Commands

Command Description
get_scoreboard Live/recent NBA scores
get_standings Standings by conference
get_teams All 30 NBA teams
get_team_roster Full roster for a team
get_team_schedule Schedule for a specific team
get_game_summary Detailed box score and scoring plays
get_leaders NBA statistical leaders
get_news NBA news articles
get_play_by_play Full play-by-play for a game
get_win_probability Win probability chart data
get_schedule Games on a specific date (no season filter; see get_team_schedule)
get_injuries Injury reports across all teams
get_transactions Recent transactions
get_futures Futures/odds markets
get_depth_chart Depth chart for a team
get_team_stats Team statistical profile
get_player_stats Player statistical profile
find_nba_player Search the NBA Stats player registry (all eras)
get_nbastats_game_log League game log via NBA Stats — history to 1946, carries NBA game ids. player_or_team="player" or player=<name or id> for player game logs
get_nbastats_player_career Career stats season by season via NBA Stats
get_nbastats_team_stats League team stats via NBA Stats — advanced ratings, pace, four factors
get_nbastats_shot_chart Per-shot court coordinates via NBA Stats
get_nbastats_play_by_play Play-by-play with coordinates for past seasons via NBA Stats
get_nbastats_advanced_boxscore Advanced box score (ratings, usage) via NBA Stats
get_live_scoreboard Real-time scores. Primary cdn.nba.com, fallback ESPN
get_live_boxscore Real-time box score. Primary cdn.nba.com, fallback ESPN
get_live_playbyplay Real-time play-by-play, most recent plays first
get_player_live_stats Real-time stats for one player in today's games

See references/api-reference.md for full parameter lists and return shapes.

Shaping Long NBA Stats Results

get_nbastats_game_log, get_nbastats_team_stats and get_nbastats_shot_chart accept sort_by, descending, limit and fields, so you can ask for the rows and columns you need instead of a full season:

# Miami's five highest-scoring games, four columns each
sports-skills nba get_nbastats_game_log --season=2025 --team=MIA \
  --sort_by=pts --limit=5 --fields=wl,pts
  • sort_by: one column to sort by. Numbers (and numeric strings) sort numerically; missing values always go last.
  • descending: true (default) or false; only used with sort_by.
  • limit: positive integer, applied after sorting.
  • fields: comma-separated keep-list. The identity columns below and the sort_by column are always kept.
  • An unknown sort_by/fields column returns an error listing the valid columns.
  • With any of these set, the response adds total_rows (matching rows before limit) and returned_rows. With none set, output is unchanged. They are applied after the fetch, so they never change the upstream request or its replay entry.
Command Always kept by fields
get_nbastats_game_log game_id, game_date, team_abbreviation, matchup
get_nbastats_game_log (player rows) game_id, game_date, player_id, team_abbreviation, matchup
get_nbastats_team_stats team_id, team_name, team_abbreviation
get_nbastats_shot_chart game_id, game_date, period

Column names are NBA.com's, lowercased (pts, fg3m, plus_minus, shot_distance).

Using ESPN and NBA Stats Together

The get_nbastats_* commands read stats.nba.com — the analytics layer (advanced ratings, shot coordinates, deep history) that ESPN's endpoints do not carry. The two sources use unrelated id systems:

  • Game ids. NBA Stats uses 10-digit ids ("0022400061"); ESPN uses event ids ("401704627"). There is no shared column — join on the game date plus the two team abbreviations.
  • Team abbreviations. Six teams are spelled differently: ESPN GS/NO/NY/ SA/UTAH/WSH vs NBA GSW/NOP/NYK/SAS/UTA/WAS. Every get_nbastats_* team filter accepts either spelling, and result rows carry both (team_abbreviation and team_abbreviation_espn).
  • Player ids. NBA person ids ("203999") and ESPN athlete ids are unrelated. Resolve names with find_nba_player; ASCII spellings match accented names ("jokic" finds "Nikola Jokić").
  • Seasons. Pass the starting year (season=2024 means 2024-25). The NBA form ("2024-25") is also accepted.
  • Dates. game_date is YYYY-MM-DD in game logs and shot charts alike.

stats.nba.com throttles by client and volume: heavy bursts (and many datacenter/cloud IPs) get silently tarpitted rather than refused. The commands fail fast with an explanatory error when that happens — wait before retrying; do not hammer. Responses are cached, and the ESPN-backed and get_live_* commands are unaffected.

Examples

Example 1: Today's scores User says: "What are today's NBA scores?" Actions:

  1. Call get_scoreboard() Result: All live and recent NBA games with scores and status

Example 2: Conference standings User says: "Show me the Western Conference standings" Actions:

  1. Derive season year from currentDate
  2. Call get_standings(season=<derived_year>)
  3. Filter results for Western Conference Result: Western Conference standings table with W-L, PCT, GB per team

Example 3: Team roster User says: "Who's on the Lakers roster?" Actions:

  1. Call get_team_roster(team_id="13") Result: Full Lakers roster with name, position, jersey number, height, weight

Example 4: Game box score User says: "Show me the full box score for last night's Celtics game" Actions:

  1. Call get_scoreboard(date="<yesterday>") to find the event_id
  2. Call get_game_summary(event_id=<id>) for full box score Result: Complete box score with per-player stats and scoring plays

Example 5: Injury report User says: "Who's injured on the Lakers?" Actions:

  1. Call get_injuries()
  2. Filter results for Los Angeles Lakers (team_id=13) Result: Lakers injury list with player name, position, status, and injury type

Example 6: Player statistics User says: "Show me LeBron's stats this season" Actions:

  1. Derive season year from currentDate
  2. Call get_player_stats(player_id="1966", season_year=<derived_year>) Result: Season stats by category with value, rank, and per-game averages

Commands that DO NOT exist — never call these

  • get_odds / get_betting_odds — not available. For prediction market odds, use the polymarket or kalshi skill.
  • search_teams — does not exist. Use get_teams instead.
  • get_box_score — does not exist. Use get_game_summary instead.
  • get_player_ratings — does not exist. Use get_player_stats instead.

If a command is not listed in the Commands table above, it does not exist.

Error Handling

When a command fails, do not surface raw errors to the user. Instead:

  1. Catch silently and try alternatives
  2. If team name given instead of ID, use get_teams to find the ID first
  3. Only report failure with a clean message after exhausting alternatives

Troubleshooting

Error: sports-skills command not found Cause: Package not installed Solution: Run pip install sports-skills

Error: Team not found by ID Cause: Wrong or outdated ESPN team ID used Solution: Call get_teams to get the current list of all 30 NBA teams with their IDs

Error: No data returned for a future game Cause: ESPN only returns data for completed or in-progress games Solution: Use get_schedule to see upcoming game details; get_scoreboard only covers active/recent games

Error: Offseason — scoreboard returns 0 events Cause: No games scheduled during the offseason (July–September) Solution: Use get_standings or get_news instead; use get_schedule to find when the season resumes

Related skills

researchmattpocock575KInvestigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.paper-context-resolverlllllllama451KRigor Paper Context helper for README-first deep learning repo reproduction. Use only when the README and repository files leave a narrow reproduction-critical gap and the task is to resolve a specific paper detail such as dataset split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumption from primary paper sources while recording conflicts. Do not use for general paper summary, repo scanning, environment setup, command execution, title-only paper lookup, or replacing Renv-and-assets-bootstraplllllllama450KRigor Setup skill for README-first deep learning repo reproduction. Use when the task is specifically to prepare a conservative conda-first environment, checkpoint and dataset path assumptions, cache location hints, and setup notes before any run on a README-documented repository. Do not use for repo scanning, full orchestration, paper interpretation, final run reporting, or generic environment setup that is not tied to a specific reproduction target.ai-research-explorelllllllama311KRigor Explore compatible skill slug for meaningful and potentially novel deep learning research candidates. Use when the researcher has chosen the task family, dataset, benchmark, evaluation method, provided SOTA references, and wants candidate-only exploration on top of `current_research` with auditable repo understanding, idea gating, fair comparison, and governed experiments written to `explore_outputs/`. Do not use for README-first trusted reproduction, open-ended direction finding, narrow c

Search skills and MCP servers

Fuzzy search across 23,137 skills and servers