An MCP server that provides access to Valorant rank history data and game statistics calculations
This server exhibits severe quality gaps across naming, descriptions, and schemas. Tools use placeholder syntax {USERNAME} in names, making them non-functional and ambiguous. Descriptions are templated and generic. Parameter descriptions are present but terse. No output schema documentation. Error handling is minimal. This server would not pass code review for production.
Get ${USERNAME}'s Latest Valorant Rank History. This is ${USERNAME}'s Valorant rank information.
Get ${USERNAME}'s latest Valorant rank history from last evaluated keys for another page of data. This is ${USERNAME}'s Valorant rank information.
Updates ${USERNAME} Valorant rank history from latest game data from Riot ID. If successful, Valorant rank history will be up to date
Calculates Valorant's Average Combat Score (ACS), a key metric that measures a player's overall impact in matches and their contribution toward Rank Rating (RR) gains or losses.
Calculates Valorant's headshot percentage, which highlights a player's aiming precision since headshots deal increased damage. However, a high headshot percentage may be less impactful if the player frequently misses non-head shots.
Tool names use {USERNAME} placeholder syntax instead of actual usernames. This makes tools non-functional for agents, the LLM cannot call a tool named 'get_{USERNAME}_valorant_rank_history' because it contains a literal brace character. Tool names must be valid identifiers known at registration time.
Descriptions replicate the {USERNAME} placeholder from tool names, making them templated and unhelpful. E.g. 'Get {USERNAME}'s Latest Valorant Rank History. This is {USERNAME}'s Valorant rank information.', the repeated placeholder is confusing and doesn't explain when/why to call this tool vs. the 'get_more' variant.
No output schema documented. The code returns ValorantRankApi objects with nested rank_history arrays and last_eval_keys, but the tool registration does not declare the output structure. LLMs cannot plan downstream calls without knowing what fields to expect.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 0 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 24 | - | v1 |
Pagination parameters in get_more_* tool are error-prone. lastEvalPuuidKey has a hardcoded length constraint of exactly 36 characters, but no explanation of what it is or how to obtain it. The error message 'The last evalated puuid key' contains a typo ('evalated' instead of 'evaluated').
Error handling returns plain strings with no guidance on recovery. Example: 'Unknown error for updating Valorant rank history.' An LLM cannot determine if this is retryable, if the API key is wrong, or if the service is down. Errors should categorize failure mode and suggest next steps.
API_KEY and VALORANT_API_BASE are stored as environment variables but used directly in HTTP requests. While not exposed as tool parameters (good), there is no validation that these are configured. The error message 'The API key setup for server is not configure' is grammatically incorrect and could be clearer.
Tools 1-3 are tightly bound to a single username (resolved from environment). This prevents a multi-user scenario where one agent instance serves multiple players. Consider redesigning to accept username as a parameter, or document this limitation clearly.