Skip to main content
Glama

audit_taste

Read-onlyIdempotent

Judges a live URL, HTML, or copy block against a taste profile, verifying design notes, banned words, contrast, and tap targets. Returns a PASS/WARN/BLOCK verdict with concrete evidence so taste violations do not ship.

Instructions

Judge a target against a taste profile. Pass html (static page/CSS), text (a copy block), or url (rendered headless; also runs delegated WCAG-contrast/tap-target measurements for owner:raven rules). Pass source_text to deterministically verify that the target's visible text remains verbatim through a content port. owner:taste rules run deterministic detectors — gradients, glow/neon (large-blur colored shadows), second accent hue, banned-word lists from the rule's negative prompt; clauses with no deterministic detector are reported honestly under not_assessed instead of guessed. owner:raven rules route through Raven's existing audit engines (page checks, contrast, tap targets) and fold results in under the delegating rule_id. Every finding cites an existing rule_id + concrete evidence — the engine prefers silence over a speculative nit. accept-verdict corpus precedents suppress previously-approved patterns. When the resolved binding carries design_notes, audit_taste VERIFIES each note against the artifact instead of only echoing it: url mode measures the rendered page's traits (scheme/luminance, canvas+WebGL, animations, scroll effects, text density, fonts, heading scale, loader, backdrop-filter), html mode extracts what it can statically, and every note comes back in note_assessments as present/partial/missing/unverifiable with trait-number evidence — design_notes are ACCEPTANCE CRITERIA for a build, not mood words. Missing notes become fidelity_findings (NOTE-, warn — block when a named library like three.js/gsap/lottie/anime.js or a branded loader is wholly absent), the target is compared against the binding's captured references (REF-* deltas on scheme, density, motion, type scale), and sparse-and-empty pages are flagged (TASTE-restraint-earned: sparseness must be earned by craft density, not achieved by deletion). fidelity_findings count toward the verdict. When a note names an expensive technique (three.js/WebGL, GSAP scroll choreography, anime.js staggered motion, glassmorphism, a branded loader, lottie, kinetic display type…), the result carries build_hints — a concrete recipe + canonical public example sources for that technique, so a failing audit hands the fix ammunition next to the missing finding; an expensive note is never license to drop it. Rules may carry a scope (e.g. portfolio-monochrome); pass surface to say what you're judging — scoped rules run at full severity on a matching surface, are skipped (reported under skipped_out_of_scope) on a non-matching one, and can warn but never block when surface is omitted. Better: pass project (or audit a bound url host) so a saved surface binding supplies the surface, per-rule overrides, and voice note automatically — on a NEW project with no binding, run get_taste_interview first (results carry a calibration_hint when calibration is missing). Verdict: BLOCK (any block finding) / WARN (any warn) / PASS.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
urlNoLive URL — rendered headless with scroll-settle; enables delegated contrast/tap-target measurement.
htmlNoFull HTML/CSS of the page to judge.
textNoA copy/text block to judge (voice/banned-word rules).
profileYesTaste profile name (see list_taste_profiles).
projectNoProject identifier — resolves a saved surface binding (see get_taste_interview / bind_taste_surface) that supplies the surface and per-rule overrides automatically. url-mode audits also resolve bindings by hostname.
surfaceNoWhat surface is being judged (e.g. 'portfolio', 'product-site', 'deck') — activates/skips scope-tagged rules by token match. Omit if unsure: scoped rules then warn instead of block.
source_textNoOriginal source copy for a deterministic word-token fidelity diff against the target's visible text.
document_kindNo'artifact' (default): the target is a build OF the surface — design_notes bind it as acceptance criteria (note_assessments/fidelity_findings run). 'portrait': the target is a document ABOUT the surface (e.g. generate_taste_portrait output) — note-fidelity is skipped and the result announces it in note_fidelity_skipped. Profile rules run in full either way.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv2.2.9

TDQS

A4.6/5.0
Behavior5/5

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

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint: false, but the description adds far more: not_assessed instead of guessing, rule delegation to Raven engines, silence-over-speculation policy, accepted contract corpus precedents, design-note verification as acceptance criteria, and build_hints on expensive techniques. The description and annotations align; there is no contradiction.

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?

The definition is very long and runs as one dense paragraph, so it is not structurally graceful. However, nearly every sentence carries a distinct behavioral policy or edge case that matters, and the opening line immediately gives the core function. The density is mostly earned rather than padded.

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?

For an 8-parameter tool with no output schema, the description is unusually complete. It covers target variants, scope resolution, project binding prerequisites, document_kind behavior, verdict semantics, finding types, fidelity_findings, note assessments, REF-* deltas, restraint findings, and build_hints. An agent has nearly everything needed to select and invoke the tool correctly.

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

Parameters4/5

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

Schema coverage is 100%, so the baseline is 3, but the description meaningfully supplements the schemas. It explains url's rendered headless scroll-settle behavior, surface omission as warn-not-block, project binding resolution, source_text as a verbatim token diff, and document_text's artifact/portrait distinction for note fidelity. These are behavioral semantics the schema alone cannot convey.

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?

Opens with a specific, non-tautological purpose: 'Judge a target against a taste profile'. It also names the concrete input modes (html, text, url) and the meaningful output dimensions, which separates it from the many generic audit_* siblings. The tool is clearly associated with taste profiles rather than generic page-score or checklist audits.

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

Usage Guidelines4/5

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

Gives strong selection and run-order guidance: pass url/html/text depending on target type, pass surface to control scope-tagged rules, pass project to reuse a saved surface binding, and run get_taste_interview first for a new project with no binding. It does not explicitly exclude sibling audits, so an agent still has to infer when not to use this tool in favor of, for example, audit_page or audit_content.

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

Install Server

Other Tools