VALORANT reference data MCP (valorant-api.com). Static game reference data — agents, weapons, maps. Keyless.
Server provides 4 well-named, read-only tools with clear descriptions and basic input schemas. Naming follows verb_noun convention (list_*, get_*). Descriptions are concise (60 - 180 chars) and explain purpose. However, parameter descriptions are minimal or absent for optional params, output schemas are undocumented, and error handling lacks recovery guidance. No pagination support despite list tools. Missing tool annotations (readOnlyHint). Baseline: 194 chars avg description, 72 chars avg param annotation, this server averages ~100 chars description but ~30 chars param annotation.
VALORANT reference data (valorant-api.com) — full details for one agent by uuid: description, role, role description, and all abilities. Static game data, keyless.
VALORANT reference data (valorant-api.com) — list playable agents with their roles and abilities. Static game data, keyless. Optionally filter by role.
VALORANT reference data (valorant-api.com) — list maps with tactical description and icon. Static game data, keyless.
VALORANT reference data (valorant-api.com) — list weapons with category, cost, fire-rate, and magazine size. Static game data, keyless.
Optional parameter 'role' in list_agents lacks type constraint (enum). Description mentions examples but does not enforce valid values. LLMs may hallucinate invalid role names.
Output schemas are not documented. Callers cannot predict response structure (e.g., does list_agents return 'agents' array or 'data'? What fields per agent?). LLMs must infer from code.
No pagination support (limit, offset, page_size, next_cursor). list_weapons and list_maps return all results unbounded. If API grows, context window exhaustion risk.
Error responses are generic ('agent not found', 'Unknown tool'). No recovery guidance. If get_agent fails, LLM does not know to call list_agents first to find valid uuid.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 66 | 2026-07-28+ | v2 |
Tool annotations missing. No readOnlyHint, destructiveHint, or idempotentHint declared. Clients cannot infer that all tools are safe read-only operations.