Skip to main content
Glama

Delimit Design Audit Render

delimit_design_audit_render

Audit rendered web pages in a real browser per viewport to catch overflow, empty placeholders, tiny text, low contrast, and visible errors before deploy.

Instructions

Audit rendered pages in a real browser the way a person sees them.

When to use: after building or changing a website, before deploy, to catch overflow, one-item desktop screens, empty placeholders, tiny text, low contrast and visible error states. When NOT to use: to lint CSS source without a running site (use delimit_design_validate_responsive).

Sibling contrast: delimit_design_validate_responsive reads source files; this renders each page in headless Chromium per viewport and colour scheme and measures the result.

Side effects: fetches the URLs (network), writes PNG screenshots and summary.json to output_dir via backends.tools_design_render.audit in a subprocess. Returns status "unavailable" with install steps when Playwright or Chromium is missing; never raises.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "urls": "<urls>"
    },
    "name": "delimit_design_audit_render"
  }
}

Input Schema

NameRequiredDescriptionDefault
urlsYesPage URLs to audit (http/https or a local HTML file), as a list or comma string. Required.
full_pageNoScreenshot the whole page instead of the first screen.
selectorsNoOptional CSS selectors, e.g. {"item": ".article-card"} names the list item counted in the first screen.
viewportsNoViewport sizes as "WIDTHxHEIGHT" list or comma string. None = 390x844, 768x1024, 1366x768, 1440x900, 1920x1032.
output_dirNoDirectory for screenshots and summary.json. None = ~/.delimit/design_audits/<timestamp>.
thresholdsNoOptional threshold overrides (e.g. {"dominant_font_min_px": 14, "downgrade": ["contrast"]}).
color_schemesNoprefers-color-scheme values to render: "light", "dark", "no-preference". None = light and dark.

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv4.25.0

TDQS

A4.2/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description goes well beyond the annotations, which only declare readOnlyHint=false and destructiveHint=false: it discloses network fetches of the URLs, PNG and summary.json writes into output_dir, subprocess execution, a degraded "unavailable" return when Playwright/Chromium is absent, and a no-raise contract. These are the exact behavioral facts an agent needs before invoking a tool that touches the network and disk.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the core action, then cleanly chunked into when-to-use, when-not, sibling contrast, and side effects. Nearly every line earns its place. The only marginal content is the internal module path "backends.tools_design_render.audit," which tells the agent little it can act on, though "in a subprocess" does carry real isolation meaning.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given seven parameters, a mutation with filesystem and network side effects, and an available output schema (so return values need not be explained), the description covers the remaining gaps: isolation, artifacts produced, and degraded-mode behavior. An agent knows what it gets, where artifacts land, and what failure looks like.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so every one of the seven parameters is already documented inline, including the default viewport set and the selectors example. The description adds no parameter-level detail beyond the schema (no threshold key list, no URL format nuance), so the baseline 3 applies — the schema does the heavy lifting here.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Uses a specific verb+resource ("Audit rendered pages in a real browser") and immediately enumerates the defect classes it targets: overflow, one-item desktop screens, empty placeholders, tiny text, low contrast, visible error states. It further distinguishes itself from the nearest sibling by stating that validate_responsive reads source files while this renders each page in headless Chromium. An agent can select this tool without opening any schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicit "When to use" (after building or changing a site, before deploy) and "When NOT to use" (to lint CSS source without a running site) with the alternative named directly. The exclusion names a concrete precondition — no running site — that makes the sibling the correct choice, leaving nothing to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools