Single tool with excellent naming, comprehensive description, and complete input schema. The tool name 'align_shapes_to_reference_slide' clearly conveys action (verb_noun pattern). Description is exceptionally detailed (1,200+ chars) with WHAT/WHEN/HOW sections, usage examples, and shape type documentation. Input schema is fully typed with proper enums for shape types and clear parameter descriptions. However, there is no visible output schema documentation in the provided source, preventing a higher score. Error handling and recovery guidance are absent from the description.
Copy the position of one or more shapes (title, subtitle, footnote) from a reference slide and apply to other slides. WHAT IT DOES: This tool takes a reference slide number and copies the exact positions (left, top, width, height coordinates) of specified shapes to multiple target slides. You can align multiple shape types in a single operation. WHEN TO USE: - User says: "Make all titles match slide 2" or "align titles to slide 2" - User says: "Make slide 5 match slide 1's layout" (copy multiple shapes) - User says: "Align titles and subtitles to slide 3" - User says: "Fix the footnotes on slides 4-7 to match slide 2" - User wants consistent positioning across slides - User wants to standardize layout using a reference slide HOW IT WORKS: 1. Gets coordinates of specified shapes from reference slide 2. Applies same coordinates to all target slides 3. Returns success/failure for each slide and shape type SHAPE TYPES: - "title": Main title placeholder - "subtitle": Subtitle placeholder - "footnote": Text box in bottom 15% of slide (disclaimers, citations) PARAMETERS: - reference_slide_number: Which slide to copy FROM (e.g., 2 means "use slide 2 as template") - target_slide_numbers: Which slides to apply TO (e.g., [1,3,4,5] means "update these slides") - shapes_to_align: Array of shape types to copy (e.g., ["title", "subtitle"]) EXAMPLES: Query: "Align all titles to match slide 2's position" → reference_slide_number: 2 → target_slide_numbers: [1, 3, 4, 5, 6, 7, 8] → shapes_to_align: ["title"] Query: "Make slides 5-10 match slide 1's title and subtitle layout" → reference_slide_number: 1 → target_slide_numbers: [5, 6, 7, 8, 9, 10] → shapes_to_align: ["title", "subtitle"] Query: "Fix everything on slide 7 to match slide 4" → reference_slide_number: 4 → target_slide_numbers: [7] → shapes_to_align: ["title", "subtitle", "footnote"] Query: "Standardize footnotes using slide 3 as reference" → reference_slide_number: 3 → target_slide_numbers: [1, 2, 4, 5, 6] → shapes_to_align: ["footnote"] IMPORTANT: Use 1-based indexing (slide 1 = first slide, not 0).
Output schema not documented. The tool description explains input parameters thoroughly but does not document what the response will contain, success/failure per slide, returned coordinates, or error structure.
No error handling guidance in tool description. If a reference slide does not exist, a shape is not found on a target slide, or coordinates cannot be applied, the description does not tell the LLM what to do next or how to recover.
No per-item failure reporting documented. For batch operations across multiple target slides, the tool should return per-slide success/failure so the LLM knows which slides were updated and which failed.
No idempotency guidance. The description does not indicate whether calling this tool twice with the same reference and target slides is safe (idempotent) or produces side effects. Critical for agent retry patterns.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | B | 77 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 48 | - | v1 |