elementor/evaluate-design runs every registered Elementor design audit over a page or subtree and returns an aggregate score with findings and recommendations.
The aggregation is real. What it aggregates is not yet.
At a glance
| Ability | elementor/evaluate-design |
| Toolset | Elementor — toolset/elementor |
| Group | Elementor › Design Audit |
| Requires | Elementor, active on the site |
| Capability | manage_options |
| Safety | Read-only — changes nothing |
| Repeatable | Idempotent — running it twice leaves the same result |
| Ships in | AcrossAI Abilities Manager 0.0.34 and later |
How it works
The individual design audits this aggregates are currently registered as skeletons: they return an empty findings list and a score of 100 rather than analysing anything. The ability works, the plumbing is real, and the analysis heuristics are not written yet — so a clean result here means “nothing was checked”, not “nothing is wrong”.
That is worth stating plainly rather than describing the intended behaviour as though it existed. An aggregate score of 100 from this ability today means the audits ran and had nothing to say, because none of them look at anything.
The structure around it is finished: the registration, the subtree scoping, the response envelope, the aggregation of findings and scores across audits. Filling in an individual audit means implementing one analysis function, and the rest already works.
The audits it will aggregate cover composition and repetition — column balance and necessity, section rivalry, emphasis drift, generic layout patterns, opportunities to use a native widget instead of a hand-built imitation.
Until those are implemented, treat this as plumbing rather than as a design opinion.
Input
| Name | Type | Required | What it is |
|---|---|---|---|
post_id | integer | Yes | The post to evaluate. |
subtree_id | string | No | Restrict the evaluation to one element and its children. |
What comes back
| Name | Type | What it is |
|---|---|---|
score | integer | The aggregate score. |
findings | array | Findings across all audits. Currently empty. |
recommendations | array | Recommendations across all audits. Currently empty. |
success | boolean | Whether the call completed. |
message | string | A one-line summary of what happened, suitable for showing a human. |
error_code | string | Present only on failure — a stable machine-readable reason, such as a missing field, a wrong field type or an unmet confirmation. |
Example
Ask your assistant:
Evaluate the design of page 412.
It resolves to one call:
{
"ability": "elementor/evaluate-design",
"parameters": {
"post_id": 412
}
}Worth knowing
- The individual audits are registered skeletons. A clean result means nothing was checked.
- The aggregation, scoping and envelope are implemented.
- Do not read a score of 100 as a design verdict.
- Read-only either way.
Related abilities
- Suggest Elementor Design Fixes — the same caveat applies
- Get Elementor Style Guide — a real read about a site’s design
- Get Elementor Pattern Guidance — guidance that is actually populated
Get the plugins:
AcrossAI Abilities Manager · AcrossAI MCP Manager
Browse the rest: Every ability, by category