Hudl Volleyball MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_user_teamsA | Discover and list all Hudl teams associated with your account to find your Team ID. Retrieves team names, team IDs, club/school organizations, roles (e.g. Coach), and active seasons directly from your authenticated Hudl account. Args: output_format: Format of the output: 'markdown' (clean table) or 'compact' (token-efficient JSON). force_refresh: If True, bypasses local 15-minute disk cache and forces a live fetch. Returns: A formatted Markdown table or compact JSON dictionary of your teams. |
| list_volleyball_matchesA | Fetch recent volleyball matches for a team from Hudl. Returns a table or compact JSON containing match IDs, dates, opponent names, results, and set scores. Call this tool first when a user asks about match results or team performance to discover the appropriate match_id for downstream analysis. Args: team_id: Optional Hudl team ID. If omitted, uses auto-detected team ID. limit: Maximum number of recent matches to retrieve (default: 10). output_format: Output format: 'markdown' (table) or 'compact' (token-efficient JSON). force_refresh: If True, bypasses local 15-minute disk cache and forces a live fetch. Returns: A formatted table or compact JSON of matches or an actionable error message. |
| get_match_player_statsA | Fetch individual athlete box scores and volleyball statistics for a match. Retrieves player stats, strips unnecessary UI metadata, ensures hitting percentage ((Kills - Errors) / Total Attacks) is accurately calculated, and formats the output into a clean, standard volleyball box score table or compact JSON including Kills (K), Attack Errors (E), Total Attacks (TA), Hitting % (Hit%), Assists (A), Service Aces (SA), Service Errors (SE), Digs (D), Block Solos (BS), Block Assists (BA), Total Blocks (TB), and Total Points (PTS). Args: match_id: The unique Hudl match identifier (obtainable via list_volleyball_matches). team_id: Optional Hudl team ID to filter stats to a specific team. output_format: Output format: 'markdown' (standard box score table) or 'compact' (token-efficient JSON). force_refresh: If True, bypasses local 7-day disk cache and forces a live fetch. Returns: A formatted volleyball box score table or compact JSON. |
| get_rotation_breakdownA | Fetch tactical performance metrics broken down by volleyball rotation (P1 through P6). Returns sideout efficiency (earned sideout points / total receive opportunities), point scoring percentage while serving, first-ball sideout (FBSO) conversion rates, and setter distribution tendencies for each rotation (P1: Setter in Zone 1 serving/receiving, P2: Setter in Zone 6, P3: Setter in Zone 5, P4: Setter in Zone 4, P5: Setter in Zone 3, P6: Setter in Zone 2). Args: match_id: The unique Hudl match identifier (obtainable via list_volleyball_matches). team_id: Optional Hudl team ID to filter breakdown to a specific team. output_format: Output format: 'markdown' (table) or 'compact' (token-efficient JSON). force_refresh: If True, bypasses local 7-day disk cache and forces a live fetch. Returns: A formatted table or compact JSON of rotation analytics (P1 through P6). |
| get_game_actionsA | Retrieve detailed tagged volleyball actions and touch events from a match/video breakdown. Returns an event-by-event timeline including set number, score, video timestamp, player jersey/name, action type (serve, receive, set, attack, dig, block), result/quality (kill, error, ace, assist, in-play, rating), and tactical details (start/end zones, attack routes, blocker count, FBSO). Supports DataVolley (.dvw) skill codes (S, R, E, A, B, D, F) and evaluation codes (# Perfect/Kill/Ace, + Positive, ! Medium, - Poor, = Error). Args: match_id: The unique match identifier, Balltime video ID, Hudl video hex ID, or match name/title. set_number: Optional set number (e.g. 1, 2, 3) to filter events to a specific set. action_type: Optional action filter: 'attack', 'serve', 'receive'/'pass', 'set', 'dig', 'block'. player: Optional player filter: jersey number (e.g. '17', '1') or player name. team: Optional team filter: 'a', 'b', or team name (e.g. 'Lutry' or 'Voléro'). quality: Optional quality/result filter: 'kill', 'error', 'ace', 'assist', 'in_play'. limit: Maximum number of events to return (default: 50). offset: Offset for paging through match events (default: 0). output_format: Output format: 'markdown' (table) or 'compact' (token-efficient JSON). force_refresh: If True, bypasses local 7-day disk cache and forces a live fetch. Returns: A formatted table or compact JSON of tagged match actions and event statistics. |
| login_to_hudlA | Open an interactive pop-up browser window to log into Hudl and save session state. Opens a visible browser to https://www.hudl.com/login where you can log in normally using your password manager (1Password, Bitwarden, iCloud Keychain, etc.), Google/Apple SSO, or two-factor authentication (2FA). The moment navigation to https://www.hudl.com/home is detected, the session state is automatically saved to ~/.hudl_mcp/auth_state.json, the browser closes, and your active session is updated without ever storing your password. Args: timeout_seconds: Maximum time in seconds to wait for login completion (default: 300 = 5 minutes). Returns: A confirmation status message when login completes or details on any error. |
| refresh_hudl_cookieA | Refresh the Hudl session authentication cookie. If credentials are configured in .env, attempts automated refresh. Otherwise, recommends or triggers interactive pop-up browser login via login_to_hudl. Returns: A confirmation status message indicating success or details on how to log in. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| scout_opponent | Generate a comprehensive tactical scouting report on an opponent team. This workflow guides the AI through locating recent matches against or featuring the opponent, analyzing top scoring threats, identifying serve-receive weaknesses, and designing a match game plan. |
| analyze_rotation_weaknesses | Diagnose sideout and point scoring vulnerabilities across rotations P1 through P6. This workflow evaluates rotation-by-rotation performance to isolate tactical flaws in serve-receive formations, setter distribution, or transition scoring. |
| match_mvp_breakdown | Determine the Most Valuable Player (MVP) of a match using advanced volleyball metrics. Calculates True Point Scoring (Points = Kills + Aces + Blocks), Efficiency ((K - E) / TA), and Defensive Impact (Digs, Passer Rating) to select the MVP with quantitative rigor. |
| datavolley_export | Generate standardized DataVolley (.dvw) scouting codes for match touches. Translates tagged match actions into standard DataVolley syntax: [Team (* or a)][Jersey (2 digits)][Skill (S/R/E/A/B/D)][Evaluation (#/+/!/-/=)][Zones/Combo]. |
| serve_receive_audit | Conduct a comprehensive serve-receive and passing performance audit. Analyzes 3-point pass percentages, reception errors, setter in-system rates, and identifies opponent serving pressure points. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 7 tools
The data tools are mostly distinct: team discovery, match listing, player box scores, rotation analytics, and event-level actions each target a different output. Some overlap exists between get_match_player_stats and get_game_actions because both provide match-level performance data, but the descriptions clarify summary stats versus event timeline. The two auth tools are also distinguishable as interactive login versus cookie refresh, though refresh can trigger login.
All names use snake_case and generally follow a verb-first pattern, such as get_user_teams, list_volleyball_matches, and get_match_player_stats. Minor deviations appear with login_to_hudl and refresh_hudl_cookie, but they remain readable and predictable. The repeated 'get_' prefix for several data tools is consistent enough for agent selection.
Seven tools is well-scoped for a read-only sports analytics MCP server with authentication support. Each tool maps to a meaningful capability: team discovery, match listing, player stats, rotation analytics, event actions, login, and cookie refresh. There is no obvious padding or missing auth/helper utility.
The server covers the core lifecycle of viewing Hudl volleyball data: discover teams, list matches, inspect player box scores, analyze rotations, and drill into tagged actions. Authentication is handled with both interactive login and cookie refresh. Minor gaps remain, such as season-level aggregate stats or roster lookup, but core workflows are covered.