Four tools with clear verb-noun naming (list_*, get_*) and well-structured schemas. All tools have descriptions (10 - 200 chars, within baseline 34 - 392 range) and documented input parameters with types. Pagination support in list_diagrams (offset, limit, max 100). However, output schemas are not explicitly documented in the source, only inferred from API responses. Parameter descriptions are functional but lack dependency hints (e.g., 'Call get_diagram_sheets first if sheetCount > 3' is mentioned in get_diagram_image description but not reciprocally in get_diagram_sheets). No tool annotations (readOnlyHint, etc.) despite all being read-only. Error handling present but generic (ApiError type with kind/message/cause), no recovery guidance in tool descriptions.

• Add explicit output schema documentation to each tool description. E.g., 'Returns: { organizations: [{ id, key, name }], count }' or link to a schema reference.
• Add readOnlyHint: true to all four tools in the MCP tool registration to signal they are safe discovery/read operations.
• Enhance error descriptions with recovery paths. E.g., 'If organizationKey is invalid, call list_organizations() to discover valid keys.'
• Document parameter format constraints inline. E.g., 'organizationKey: alphanumeric string, typically 10 chars (e.g., hEkx29Hj9h)' and 'diagramId: numeric string'.
• Add min/max bounds to numeric parameters: width/height 100 - 4000, limit 1 - 100, offset ≥0.