Cricket MCP — wraps CricAPI (api.cricapi.com) for live cricket data.
Cricket MCP has clear, well-structured tool definitions with good naming conventions and solid descriptions. All 4 tools follow verb_noun patterns (current_matches, match_scores, search_players, match_info). Descriptions are substantive (100-200 chars), explaining WHAT each tool does and WHEN to use it. Input schemas are present with type definitions and required fields properly declared. However, output schemas are not documented, LLMs cannot predict response structure. Parameter descriptions are adequate but lack format constraints (e.g., offset range, search string length). Error handling returns structured {error, message} objects but lacks recovery guidance or categorization. No tool annotations (readOnlyHint, idempotentHint) despite all tools being read-only. The _apiKey parameter is a security anti-pattern, credentials should never be tool parameters.
List current and recent cricket matches (Test/ODI/T20) with live scores, teams, venue, and status. Use this for "what cricket matches are live/on right now". Returns match IDs usable with match_info.
Get full information and scorecard for a single cricket match (Test/ODI/T20) by its match ID. Returns teams, venue, innings scores, toss, and winner. Get the id from current_matches.
Get a lightweight feed of live cricket scores across current matches. Returns each match as a compact "team1 vs team2" with running scores and status. Best for quick live cricket score checks.
Search for cricket players by name. Returns player IDs, names, and country. Use this to look up a cricketer before fetching more detail.
Output schemas not documented. LLMs cannot predict response structure (fields, types, nesting). Responses are shaped but undocumented, forcing LLMs to infer structure from examples.
_apiKey exposed as tool parameter violates secret-injection pattern. Credentials in parameters leak into agent logs and prompt history. Move to server-side injection via environment variables.
No tool annotations despite all tools being read-only. Missing readOnlyHint annotation prevents agents from understanding safety properties and optimizing caching/retry strategies.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 59 | 2026-07-28+ | v2 |
Parameter constraints lack specificity. 'offset' has no min/max bounds; 'search' has no length limits. LLMs may pass absurd values (offset=-1, search with 10k chars) causing API failures.
Error responses lack recovery guidance. {error, message} structure is present but does not categorize errors as retryable/user-fixable/fatal or suggest next steps (e.g., 'Try search_players() first').