This server has 6 tools with basic descriptions and minimal parameter documentation. Tool names follow verb_noun convention (onto_load, onto_validate, onto_reason, onto_classify_el, onto_lint_file, onto_query), which is good. However, descriptions are terse (10-50 chars), parameter descriptions are sparse or missing, and input schemas lack proper type constraints. onto_classify_el has an empty input schema with no parameters documented. No output schemas are documented. Error handling is marked as present but not visible in the code sample. The server appears to be an Obsidian plugin wrapping an external ontology engine, but the MCP tool definitions lack the rigor needed for reliable LLM invocation.

• Expand all tool descriptions to 100-200 characters. Include WHAT the tool does, WHEN to use it (vs similar tools), and WHAT it returns. Example: 'onto_reason: Apply OWL-RL inference rules to the loaded graph to derive implicit facts. Use after onto_load to expand the knowledge base. Returns inferred triples as Turtle RDF.'
• Add parameter descriptions for all inputs. For onto_reason.profile, declare enum: ['owl-rl', 'owl-el', 'rdfs'] with explanation of each. For onto_load.turtle, specify: 'Valid Turtle RDF syntax; max 10MB; invalid syntax returns error with line number.'
• Document output schemas for each tool. Example for onto_query: 'Returns JSON object with keys: results (array of bindings), variables (array of SPARQL variable names), count (number of results).'
• Add error recovery guidance. Example: 'If onto_load fails with syntax error, call onto_validate first to identify the malformed line.'
• Clarify tool composition. Document that onto_load must be called before onto_reason, onto_classify_el, or onto_query. Add dependency hints in descriptions.

Inferred effective spec: <=2025-11-25.

• Relies on Stateful initialize / Mcp-Session-Id (removed; protocol is stateless) - make each request self-contained