get_preview_qrA terminal-scannable QR code for a scene. Scan it with a phone camera and the scene runs live in the Keyframer preview app on real hardware — edits sync as they happen. Public scenes (including guest scenes) work without a key; private scenes need an API key on a paid plan.
| Name | Required | Description | Default |
|---|---|---|---|
| scene_id | Yes | Scene id from create_scene or list_scenes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The readOnlyHint annotation already establishes safety, and the description adds useful behavioral detail: scanning runs the scene live, edits sync as they happen, and authentication requirements vary by scene visibility. This goes beyond the annotation without contradicting it.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three sentences with no filler: the first identifies the deliverable, the second explains the preview behavior, and the third covers authentication. The key information is front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The definition is adequate for a single-parameter, read-only tool: it describes the output, the on-device behavior, and auth requirements. There is no output schema, but 'terminal-scannable QR code' sufficiently conveys the return payload for this use case.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema provides 100% coverage for scene_id, including a helpful source hint ('from create_scene or list_scenes'). The description adds no parameter-specific meaning, but the schema already carries that burden effectively.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states that the tool produces a terminal-scannable QR code for a scene and explains what the QR code does. It is distinct from scene retrieval siblings by focusing on live device preview, though it does not explicitly contrast itself with a named alternative.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It gives actionable context: public and guest scenes work without a key, while private scenes require an API key on a paid plan. It does not explicitly name alternate tools or exclusion conditions, but the intended use case is clear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Each tool maps to a distinct resource or action: scene CRUD, validation, motion sampling, recipe lookup, code generation, and QR preview. Potential pairs like validate_scene vs sample_scene are clearly separated as structural validation vs motion evaluation, and get_scene vs get_recipe differ by source and purpose.
Tool names consistently follow a verb_noun snake_case pattern: create_scene, get_scene, list_scenes, update_scene, validate_scene, search_recipes. The few compound nouns like get_preview_qr and get_scene_format still fit the same predictable convention.
11 tools is well within the ideal range and each tool serves a clear part of the scene authoring workflow: discovery, validation, sampling, editing, compiling, and previewing. None feel redundant or extraneous for the stated purpose.
The toolset covers the core scene lifecycle well: create, read, list, update, validate, sample, generate code, and preview. The main gap is the absence of a delete_scene tool, which is a minor omission since the authoring and iteration workflow is otherwise fully supported.