Least-privilege Google Forms MCP server
Server has 22 well-structured tools with consistent naming (verb_noun pattern), complete input schemas with type constraints, and descriptions for all tools and parameters. Naming is clear and action-oriented (create_form, add_text_question, delete_item). All parameters have descriptions and type definitions. However, descriptions are brief (10-50 chars typical) and lack context about when to use each tool or what happens on error. Output schemas are not documented, LLMs cannot predict response structure. Error handling is generic ('Google Forms operation failed'). No tool annotations (readOnlyHint, destructiveHint) despite clear risk levels (WRITE, DESTRUCTIVE, READ_ONLY). Missing guidance on idempotency, retryability, and recovery paths.
Add radio, checkbox, or dropdown choices.
Add a date or date-time question.
Add an image from a public URL.
Add a radio or checkbox grid.
Compatibility alias for a radio-button question.
Add a star, heart, or thumb rating.
Add a linear numeric scale.
No output schemas documented. LLMs cannot predict response structure or plan downstream calls. Tools return dict[str, Any] with no field documentation.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite clear risk levels. LLMs cannot determine which tools are safe to retry or have side effects.
Error handling is generic ('Google Forms operation failed'). No guidance on retryability, user-fixable errors, or recovery paths. LLMs cannot self-correct.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 62 | 2026-07-28+ | v2 |
Add a section/page break.
Add non-question explanatory text.
Add a short-answer or paragraph question.
Add a time-of-day or duration question.
Add a YouTube video.
Create a form with an optional description.
Permanently delete an item by zero-based index. Confirm the exact item with the user first.
Retrieve the complete form structure.
Retrieve a single form response by ID.
List form responses with optional filtering and pagination.
Move an item between zero-based positions.
Set quiz mode and/or email collection.
Publish/unpublish and enable/disable responses. Confirm with the user before changing availability.
Set quiz points, answer key, and optional right/wrong feedback for a question item.
Update the visible form title and/or description.
Descriptions are brief (10-50 chars) and lack context. Missing 'when to use' guidance, prerequisites, and dependency hints. Baseline is 194 chars for A+ tools.
list_responses lacks pagination documentation. No mention of result limits, total count, or next_cursor behavior. Large result sets could exhaust context.