Accepted a fully specified HTML/CSS/JavaScript animation prompt immediately, but still misread part of the request as interactive webpage behavior rather than animation-first motion graphics.
What was measured
Input Handling
Accepts plain-language prompts without detailed specs.
decisive for this rankingcapability
The ranking is specifically about text inputs, so a tool must handle plain-language prompts without heavy specification to do the job well. (2 of 3 judges)
What was given, what came back
Test input: Search engines basics explainer · text
Input — what we sent
The exact prompt
Create an animation video explaining how search engines work.
A vague, linear educational animation prompt asking the tool to explain how search engines work with minimal guidance. Designed to stress inference, scene structuring, pacing, and coherent visual flow from an underspecified request.
Output — unretouched
Grok — Screen Recording 2026-05-04 135441.mp4
Also checked on this input — same tool, 4 other criteria
Code Generation Quality✓ WorkedProduced syntactically correct HTML/CSS/JavaScript code on the first pass.Output Quality⚠ StruggledThe output read more like an interactive webpage than an animation video: it included interactive elements despite the no-interaction requirement, had cramped spacing and overlapping layout issues, and lacked clear visual hierarchy.Output Quality✗ FailedProduces output that reads more like an interactive webpage than a video-like animation, with cramped spacing, overlapping elements, weak visual hierarchy, and visible controls.Output Quality✗ FailedProduces a cluttered, webpage-like result with overlapping elements, cramped spacing, and no clear motion-graphics hierarchy.
Provenance
- Observation
- 2312144f-75e5-45c5-91b0-3f14ad09a663
- Evidence run
- c751a76b-a105-4b3e-9ff3-6c4d6adc9f04
- Study
- Generate Code-based Animations from text inputs
- Research task
- 86b9nnzmy
- Tested at
- Jun 24, 2026
- Source
- first-party
- Evidence state
- observed
- Proof shown
- input + output shown
- Cost / latency
- not captured
- Repeat run
- not captured
- Tester
- not captured
The last three rows are honest blanks, not placeholders — our capture has no field for them yet.
Query this
get_evidence({
tool: "grok"
})MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, same check — 3 other tools
measured on Input Handling
Antigravity✓ WorkedAccepts a vague educational prompt and infers a coherent linear search-engine workflow without needing detailed specs.Gemini Canvas✓ WorkedAccepts a plain-language animation prompt and infers the full search workflow without needing a detailed implementation spec.RemotionVideo✓ WorkedAccepts a plain-language explainer prompt and infers a coherent 3-stage structure without needing detailed specs.
Real inputs and real outputs, no retouching · every cell queryable via API & MCP · aidemos.com