MCP server for managing Google Forms — quizzes, grading, images, responses. Part of likezara™ MCP suite.
Server has 9 tools with complete input schemas and descriptions. Naming follows verb_noun convention (create_, rename_, set_, get_, list_, add_, batch_add_). Descriptions are present but brief (avg 80 chars, below 194-char baseline). Parameters are typed and mostly described, but lack constraints (enums, ranges, patterns). Output schemas are not documented, LLMs cannot predict response structure. Error handling is minimal; no recovery guidance. Missing tool annotations (readOnlyHint, destructiveHint). No pagination support despite batch operations.
Append (or insert at index) a True/False question with answer key.
Report whether OAuth credentials and token are in place.
Add multiple True/False questions in one batch request.
Create a new Google Form in quiz mode. Returns formId and responderUri.
Fetch full form metadata, items, and settings.
Return a concise list of items: index, itemId, title, correctAnswer, points.
Output schemas not documented. LLMs cannot predict response structure for any tool. Responses from create_quiz_form, get_form, list_questions, verify_answer_keys lack field documentation.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). LLMs cannot distinguish safe reads from destructive writes. All 9 tools lack these hints.
Parameter constraints missing. 'points' (add_true_false_question, batch_add_true_false) lacks min/max bounds. 'index' lacks range documentation. No enums for boolean-like fields. Invites invalid LLM inputs.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 54 | 2026-07-28+ | v2 |
Update form's displayed title, description, and/or Drive document title.
Toggle quiz grading mode on/off.
Compare current answer keys to an expected list. Each expected item: {"index": int, "correct": "ถูก"|"ผิด"|str}. Returns {matches: [...], mismatches: [...]}.
Error handling absent. No recovery guidance, no error categorization (retryable vs fatal), no actionable error messages. _batch_update() and _get_service() can fail silently or with raw API errors.
No pagination or result limits. list_questions and get_form may return unbounded data. No limit parameter, no cursor, no total count. Large forms risk context window exhaustion.