AI-optimized MCP server for PineScript documentation and code review
Server has 4 tools with mixed quality. All tools have descriptions and input schemas are visible, but there are significant gaps in parameter descriptions, output schema documentation, and error handling guidance. Tool naming follows verb_noun convention appropriately (pinescript_reference, pinescript_review, pinescript_syntax_compatibility, get_service_version). However, parameter descriptions lack sufficient detail about constraints, formats, and expected ranges. Output schemas are not documented. The pinescript_review tool has complex conditional logic (source_type determines which parameters are required) but this dependency is not clearly explained. Overall structure is present but falls short of production-grade quality.
Retrieve the MCP server version, build information, and deployment status.
Search PineScript documentation with enhanced semantic matching and streaming support for large result sets.
Review PineScript code against style guide and language rules. Supports single files, directories, and streaming for large results via JSON chunks.
Check PineScript code for syntax compatibility issues and deprecated functions with migration guidance.
Output schemas not documented for any tool. Tools return results but LLMs cannot predict field structure, making downstream chaining difficult and forcing agents to guess at response format.
pinescript_review tool has conditional parameter requirements based on source_type enum value, but this dependency is not documented in parameter descriptions. When source_type='code', 'code' is required; when source_type='file', 'file_path' is required; when source_type='directory', 'directory_path' is required. LLMs will not understand these constraints without explicit documentation.
Parameter descriptions lack format specifications and constraints. For example, 'query' in pinescript_reference has no length limit or format specified. 'max_results' has no explanation of what happens if you exceed 100. 'version' parameters lack explanation of valid formats (is it 'v5', 'v6', '5', '6'?). 'chunk_size' lacks guidance on practical ranges.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | C | 62 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 24 | - | v1 |
get_service_version has empty input schema (properties: {}) but no description explaining what fields it returns or why an agent would call it. Description is only 55 chars and uninformative about the response structure.
No error handling guidance in any tool description. If pinescript_reference returns no results, or pinescript_review encounters a file that cannot be read, tools provide no recovery hints. Descriptions do not indicate whether tools can fail, what causes failures, or what the LLM should do next.
pinescript_reference offers 'format' parameter with enum ['json', 'stream'] but description does not explain the difference or when to use each. What does 'stream' actually return? How should the LLM handle streaming output?
pinescript_review 'severity_filter' parameter description uses generic language ('Filter violations by severity') without explaining what 'error' vs 'warning' vs 'suggestion' means in PineScript context, or what the LLM should choose for common scenarios.