A Model Context Protocol server for wxflows that exposes dynamic tools from wxflows as MCP tools
The wxflows-server exposes a single tool 'dynamic_wxflows_tools' with severe definition and usability issues. The tool name violates naming conventions (no action verb, generic 'dynamic' prefix). The input schema is critically incomplete, it defines only a 'query' parameter as a free-form string with NO type constraint, no format guidance, and no examples. The description is vague ('Dynamic tools retrieved from wxflows API...') and does not explain what GraphQL query strings are valid, what the tool actually does, or when to use it. There is no documented output schema, no pagination support, no error guidance, and no field-level documentation. The tool appears to accept arbitrary GraphQL queries, which is a major security and reliability red flag without input validation guidance. This falls well below production baselines: average production tools have 18-char names starting with verbs, 194-char descriptions, 4 parameters on average with type+description pairs, and fully documented output schemas.
Dynamic tools retrieved from wxflows API. Tools are fetched at runtime and can vary based on wxflows configuration. Each tool uses a GraphQL schema parameter.
Tool name does not start with an action verb. 'dynamic_wxflows_tools' is a noun phrase that does not convey what the tool does. Should be 'execute_graphql_query' or 'call_wxflows_api'.
Input schema is incomplete and lacks type safety. The 'query' parameter is defined as a free-form string with no description, format constraint, example, or validation rules. LLMs have no guidance on what GraphQL queries are valid.
No output schema documented. The tool description mentions 'GraphQL schema parameter' but does not specify what the tool returns, what fields the response contains, or how to use the output.
Tool description is generic and does not explain WHAT the tool does, WHEN to use it, or what prerequisites exist. 'Dynamic tools retrieved from wxflows API. Tools are fetched at runtime and can vary...' is vague about the actual GraphQL execution semantics.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 26 | 2026-07-28+ | v2 |
| 2026-04-07 | F | 34 | - | v1 |
No error handling guidance. If a GraphQL query is malformed or returns an error, there is no documented recovery path. The tool accepts arbitrary GraphQL strings, which is a security and reliability concern without input validation and error guidance.
No parameter descriptions beyond the minimal schema. The 'query' parameter needs a detailed description explaining valid GraphQL syntax, constraints, authentication requirements, and example usage.
Tool accepts free-form GraphQL queries with no documented validation. This is a high-risk pattern, LLMs can generate syntactically invalid or semantically dangerous queries (e.g., expensive aggregations, unintended mutations). Input validation rules and constraints must be documented.