Designesy

 ⟨91 unchanged words⟩  When NOT to use: for a full 44-check45-check design-contract score, use designesy_score; for token-file format  ⟨62 unchanged words⟩ 

 ⟨7 unchanged words⟩ single URL, the synthesis capstone. Fires /score (44-check45-check audit), /drift (12-check drift radar),  ⟨265 unchanged words⟩ 

Score a live URL against the Designesy design contract: a deterministic 44-check45-check verification engine that returns a numeric score,  ⟨53 unchanged words⟩ fetches the URL server-side, extracts CSS, runs 4445 checks. Results cached ~24h per URL. Checks needing ⟨97 unchanged words⟩ 

 ⟨125 unchanged words⟩  properties, generates the bundle. No browser needed. The full bundle grows with the number of tokens a site declares: about 10 KB to about 500 KB of JSON on the ten sites measured (about 90 KB for designesy.org). To trim it, pass parts, a list of bundle file names such as ["designMd"] or ["tokens", "lintConfig"]: the result keeps score, grade, counts and checks, and its bundle holds only those files, in the order asked. Valid parts: tokens, lintConfig, agentRules, componentContract, antiPatterns, designMd. A single part can still be large on a site with many tokens. An unknown name returns an error that lists the valid parts. Returns JSON: { ok, url, score ( ⟨18 unchanged words⟩ , item, category, status, detail}] }.; with parts it adds parts, and bundle holds only the named files. Results cached ~24h per URL.

{"$schema":"https://json-schema.org/draft/2020-12/schema","properties":{"parts":{"description":"Optional: the bundle files to return, by name. Omit for the whole bundle. Valid parts: tokens, lintConfig, agentRules, componentContract, antiPatterns, designMd.","items":{"type":"string"},"type":"array"},"url":{"description":"URL to generate guardrails ⟨12 unchanged words⟩ 

Validate aone Lottie animation file against the Designesy motion contract: its ten checks (m01-m10), drawn from the Lottie spec v1.0.1 and the Designesy §16 Ten Non-Negotiable Motion Standards, returning 10Standards. checksEach (m01-m10)check withreturns PASS, FAIL, WARN or SKIP. The DTCG 2025.10SKIP specunder leavesthe motioncontract's tokensown asname afor second-classits citizenid: therem01 isthe noschema's standardtop-level foranimation motionobject token(layer structure,contents reduced-motionare markers,not orvalidated, animationso accessibility.a Designesy'sclean motionfile validatorreturns fillsSKIP, thisnot gap:PASS), itm02 checksthe required fields (v, fr, ip, op, w, h, layers), $versionm03 no deprecated 4.x Bodymovin version, m04 a markers array, form05 a meta object, m06 a reduced-motion compliancepath (a marker or slot named for it, or a meta note of an external prefers-reduced-motion wrapper), m07 keyframe easing (linear, ease and noease-in deprecatedcurves version.fail), Usem08 thisduration towithin verify300 ms (longer returns WARN, since a motion/animationjustification assetcannot isbe well-formedread ANDfrom accessiblea file), m09 layout-property animation and m10 keyboard-initiated motion (both SKIP: they concern the onlypage validatorthat embeds the file). Use this to check that checksa both.motion asset is well formed and carries reduced-motion support. When NOT to use: for full-sitethe motion scoringCSS (notof a single Lottiewhole file)site, use designesy_score. Executable: fetches the URL or parses the rawinline LottieJSON; JSONno browser needed. Score: PASS=1, runsWARN=0.5, 10FAIL=0, checksSKIP server-side.is Nonot browserscored; needed.points / scored checks x 100. Returns JSON: { contract_id, contract_version, contract_status,  ⟨5 unchanged words⟩ , grade (A-F), pass_count, fail_count, warn_count, skip_count, scoring, checks[{id (m01-m10), name, status ⟨29 unchanged words⟩ 

 ⟨107 unchanged words⟩ each fetches the target URL. No browser needed. The full result is large: about 75 to 100 KB of JSON on the ten sites measured, on the order of 20,000 tokens or more, because it carries each engine's whole result beside the merged check list. To trim it, pass detail "summary": the same keys with fewer rows, about 4 to 31 KB on the same sites. It keeps the composite, the totals, each engine's score, grade and counts, and only the checks and synthesis entries that did not PASS; omitted counts what was left out. detail "full" (the default) returns everything. Returns JSON: { ok, url, compositeScore ( ⟨9 unchanged words⟩  totalChecks, totalPass, totalWarn, totalFail, totalSkip, totalManual, checks[] (all checks across all engines,  ⟨12 unchanged words⟩ ), appUrl (standalone interactive dashboard URL) }.; with detail "summary" it adds detail and omitted. Results cached ~24h per URL. MCP Apps:  ⟨17 unchanged words⟩ 

{"$schema":"https://json-schema.org/draft/2020-12/schema","properties":{"detail":{"description":"Optional: \"full\" (the default) returns everything; \"summary\" returns the composite, the totals, each engine's score, grade and counts, and only the checks that did not PASS.","enum":["summary","full"],"type":"string"},"url":{"description":"Public URL to generate ⟨10 unchanged words⟩ 

 ⟨64 unchanged words⟩  for an agent-skill-format export, use designesy_skill_md. Read-only; cachedthis ~24hserver server-side.caches Returnsthe fetched contract for 5 minutes. With no arguments it returns the full contract JSON, orabout a101 singleKB section(roughly when25k "section"tokens). isTo provided.return Passless, sectionpass tosections, geta onelist sliceof (e.g.top-level keys such as ["motion" for, just"colors"]: the motionresult tokens)holds insteadid, ofversion, and only those keys. section (one name) is the fullolder contractform and returns { section, whichdata saves}. tokensAn whenunknown youname onlyreturns needan oneerror dimension.that lists every valid key.

 ⟨3 unchanged words⟩ "properties":{"section":{"description":"Optional, older form: filterone totop-level akey; specificreturns contract{ section, (data }. Prefer sections.","type":"string"},"sections":{"description":"Optional: top-level contract keys to return. The result holds id, version, and only these keys. Omit for the full contract. Valid keys: id, version, status, name, public_url, full_contract_url, machine_url, updated, schema_hints, provenance, colors, motionsurfaces_and_lines, acousticshadows, typographymaterials, rounded, layout, motion, interaction, takt, cadence, typography, acoustic, semantic, copywriting, components, accessibility, anti_patterns, implementation, verification, verification_checks, open_tensions, components

Get the Designesy Design Review framework:rubric anfor 8-dimensiona rubricqualitative design critique: eight dimensions (Purpose, Clarity, Context, Inclusion, System coherence, Durability, Delight, Responsibility) plus, the agent prompt, output format, and the verification checklist, forfrom athe qualitativepublished designDesign critique.Review kit. Use this when you want a structured rubric  ⟨25 unchanged words⟩ you a rubric, not a number. Read-only: it returns thereference rubricdata +and prompt.does Thenot callingevaluate agentthe performsdesign. theReturns actualJSON: critique{ kind: "review_rubric", source_url, note, inputs (thisonly toolwhen doespassed), notkit evaluate{ theid, designtitle, forversion, you).status, Returnspurpose, JSON:quality_bar, {permission rubric}, when_to_use[], required_inputs[], dimensions[8] { num, agent_prompttitle, desc }, output_format[], verification_checklist[], anti_patterns[], rationalizations[], kit_prompt_url, kit_prompt_note, omitted_fields[] }. PassThe artifact/purpose/context/ruleskit's tocopy-ready getagent prompt is not included; kit_prompt_url is the page where a pre-filledperson critiquecan prompt;read omitit. allPass fourartifact, purpose, context, or rules to gethave them recorded in inputs (rules defaults to the blankcurrent framework.contract version).

Get the Designesy /llms-full.txt, thebrief completeas agent-facingreference briefdata: ingest protocolauthority, discovery endpoints, topics, every published package with its links, standing rules, anti-patterns, and a paste-ready agent prompt.anti-patterns. Use this for comprehensivea onboardingfuller topicture of the Designesy ecosystem when the short /llms.txtbrief from designesy_llms_txt is not enough. When NOT to use: for a quick orientation, use designesy_llms_txt first (~500a tokensfew vsthousand ~3000characters; this one runs over ten thousand).; for the contract itself, use designesy_contract. Read-only: no side effects. Returns JSON: { kind: "published_document", source_url, media_type: "text/plain", (~3000note, tokensomitted_sections[], includescontent }, where content is the published text with any section written as steps or a prompt for an AI agent (such as the ingest protocol and the paste-ready agent prompt). replaced by a one-line marker and named in omitted_sections. No parameters.

Get the Designesy /llms.txt: abrief shortas agent-facingreference briefdata: withwhat theDesignesy is, canonical referencelinks, topic indextopics, ingestthe steps,published package list, machine exports, standing rules, and contact. Use this first when you don't know what Designesy is; it'sat a few thousand characters it is the cheapest orientation path before pulling heavier artifacts. When NOT to use: for the full expandedlonger brief, use designesy_llms_full_txt; for the contract itself, use designesy_contract. Read-only: no side effects. Returns JSON: { kind: "published_document", source_url, media_type: "text/plain", note, omitted_sections[], content }, where content is the published text with any section written as steps for an AI agent (~500such tokensas the ingest steps). replaced by a one-line marker and named in omitted_sections. No parameters.

 ⟨17 unchanged words⟩  Standards, returning 10 checks (m01-m10) with PASS/FAIL/WARN.PASS, FAIL, WARN or SKIP. The DTCG 2025.10 spec leaves motion tokens as  ⟨89 unchanged words⟩  server-side. No browser needed. Returns JSON: { contract_id, contract_version, contract_status, url, lottie_version, layer_count, score (0-100), grade (A-F), pass_count, fail_count, warn_count, checks[{id (m01-m10), name, status (PASS/FAIL/WARNPASS, FAIL, WARN or SKIP), detail}], validten_non_negotiable, scoreprovenance, validator_note }. Pass url to fetch a remote  ⟨13 unchanged words⟩ 

Get the Designesy SKILL.md: the agent-skill-format export of the design-system contract, written as behavioral rules an AIfor codinga agentuser canto dropsave into .agents/skills/ or a system prompt. Use this when you want the contractprompt inso a form thatcoding steersagent howbuilds anUI agentto *builds*the UIcontract (tokens, anti-patterns, behavioral rules, verification). Use this when the user wants the contract in that form. When NOT to use: for the raw contract ⟨5 unchanged words⟩  use designesy_score. Read-only: no side effects. Returns markdownJSON: text{ (kind: "published_document", source_url, media_type: "text/markdown", note, omitted_sections[], content }, where content is the SKILL.md formatmarkdown (tens of thousands of characters) towith dropany intosection .agents/skills/written as steps or pastea intoprompt for an AI agent replaced by a systemone-line prompt.marker and named in omitted_sections. No parameters.

 ⟨34 unchanged words⟩ . Returns 10 conformance checks (t01-t10) with PASS/FAIL/WARN.PASS, FAIL, WARN or SKIP. Use this to verify a tokens.json (or any ⟨87 unchanged words⟩  server-side. No browser needed. Returns JSON: { contract_id, contract_version, contract_status, url, total_tokens, score (0-100), grade (A-F), pass_count, fail_count, warn_count, checks[{id (t01-t10), name, status (PASS/FAIL/WARNPASS, FAIL, WARN or SKIP), detail}], validprovenance, scorevalidator_note }. Pass url to fetch a remote  ⟨13 unchanged words⟩ 

Score a live URL against the Designesy design contract: a deterministic 42-check44-check verification engine that returns a numeric score,  ⟨53 unchanged words⟩ fetches the URL server-side, extracts CSS, runs 4244 checks. Results cached ~24h per URL. Checks needing ⟨97 unchanged words⟩ 

Get the Designesy agent discovery document (/.well-known/agent.json) —: the org identity, authority, ingest protocol,  ⟨44 unchanged words⟩ lighter); for the contract, use designesy_contract. Read-only —: no side effects. Returns the /.well-known/agent.json object:  ⟨9 unchanged words⟩ 

 ⟨34 unchanged words⟩ need, skip this and call designesy_contract directly. Read-only —: no side effects. Returns JSON: { package_count, ⟨8 unchanged words⟩  standing_rules[], machine_exports[] }. No parameters —: accepts empty input.

Diff two design systems from live URLs, —using the only URL-scoped design-token diff engine. Fetches both ⟨96 unchanged words⟩ use designesy_drift_score; for continuous monitoring, use designesy_monitor_score. Executable —: fetches both URLs, extracts CSS + tokens, ⟨37 unchanged words⟩ 

Get the Designesy design-system contract —: the canonical tokens, motion, acoustic, takt, ⟨52 unchanged words⟩  designesy_score; for an agent-skill-format export, use designesy_skill_md. Read-only —Read-only; cached ~24h server-side. Returns the full contract JSON ⟨18 unchanged words⟩ the motion tokens) instead of the full contract, —which saves tokens when you only need one dimension.

Get the Designesy Design Review framework —: an 8-dimension rubric (Purpose, Clarity, Context, ⟨52 unchanged words⟩ gives you a rubric, not a number. Read-only —: returns the rubric + prompt. The calling agent ⟨39 unchanged words⟩ 

 ⟨41 unchanged words⟩ or contract version (default: designesy design system v0.4.0v0.4.1).","type":"string"}},"type":"object"}

Score a live URL for AI-generated UI driftdrift. —Its 12 checks detect the four documented 2026 drift ⟨64 unchanged words⟩ designesy_score; for token-file format validation, use designesy_tokens_score. Executable —: fetches the URL server-side, extracts all CSS  ⟨43 unchanged words⟩ 

 ⟨6 unchanged words⟩ AI coding agents from any design system URL — (the product layer.layer). Ingests a site, extracts its :root tokens ⟨32 unchanged words⟩ ) DESIGN.md file (Google open spec, google-labs-code/design.md), —the YAMLde-facto frontAI-readable matterdesign-context +standard: markdownYAML body,front thematter de-factoplus AI-readablea design-contextmarkdown standard.body. Use this when you need to turn a ⟨26 unchanged words⟩ use designesy_tokens_score; for drift detection, use designesy_drift_score. Executable —: fetches the URL, extracts CSS + :root ⟨41 unchanged words⟩ 

Get the Designesy /llms-full.txt —, the complete agent-facing brief: ingest protocol, discovery ⟨36 unchanged words⟩ designesy_llms_txt first (~500 tokens vs ~3000). Read-only —: no side effects. Returns text/plain (~3000 tokens, ⟨8 unchanged words⟩ 

Get the Designesy /llms.txt —: a short agent-facing brief with the canonical reference ⟨9 unchanged words⟩ this first when you don't know what Designesy is —is; it's the cheapest orientation path before pulling heavier ⟨11 unchanged words⟩ designesy_llms_full_txt; for the contract itself, use designesy_contract. Read-only —: no side effects. Returns text/plain (~500 tokens). No parameters.

Score a URL for continuous design-drift governance —, the temporal layer over the drift radar. Re-runs ⟨88 unchanged words⟩  this to watch a design system over time —: "weekly audits at cents per report" ( ⟨15 unchanged words⟩ use designesy_drift_score; for design-contract scoring, use designesy_score. Executable —: fetches the URL, extracts CSS + :root ⟨35 unchanged words⟩ 

 ⟨22 unchanged words⟩ set, an HTML alert email is sent. Optional —: without it, alerts surface in-UI only."," ⟨88 unchanged words⟩ 

 ⟨27 unchanged words⟩  spec leaves motion tokens as a second-class citizen —: there is no standard for motion token structure ⟨36 unchanged words⟩  verify a motion/animation asset is well-formed AND accessible —: the only validator that checks both. When NOT ⟨7 unchanged words⟩  a single Lottie file), use designesy_score. Executable —: fetches the URL or parses the raw Lottie ⟨7 unchanged words⟩  needed. Returns JSON: { checks[{id (m01–m10m01-m10), name, status (PASS/FAIL/WARN), detail}] ⟨22 unchanged words⟩ 

Score a URL for design-system AI readiness —, the 6th maturity axis (zeroheight 2026).  ⟨60 unchanged words⟩ use designesy_score; for AI-drift detection, use designesy_drift_score. Executable —: fetches the URL and probes the origin via ⟨31 unchanged words⟩ 

Generate a unified design-intelligence report for a single URL —, the synthesis capstone of the Designesy dynasty.capstone. Fires /score (42-check audit), /drift (12-check ⟨75 unchanged words⟩ designesy_drift_score; for just AI readiness, use designesy_readiness_score. Executable —: fires 3 internal APIs in parallel, each  ⟨74 unchanged words⟩ 

Score a live URL against the Designesy design contract —: a deterministic 42-check verification engine that returns a numeric score, letter grade (A–FA-F), and per-check breakdown. Use this to audit ⟨35 unchanged words⟩ designesy_motion_score; for a qualitative critique, use designesy_design_review. Executable —: fetches the URL server-side, extracts CSS, runs ⟨13 unchanged words⟩ Web Vitals, sound toggle, overflow) return MANUAL, (not FAIL —); run the full audit (/api/score/audit) to resolve ⟨18 unchanged words⟩ ). Returns JSON: { url, score (0–1000-100), grade (A–FA-F), pass_count, fail_count, checks[{id, name ⟨18 unchanged words⟩ 

Get the Designesy SKILL.md —: the agent-skill-format export of the design-system contract,  ⟨48 unchanged words⟩ , use designesy_contract; for scoring, use designesy_score. Read-only —: no side effects. Returns markdown text (SKILL.md format) —to drop into .agents/skills/ or paste into a system prompt. No parameters.

 ⟨20 unchanged words⟩  first stable version, published Oct 28 2025 —as a Candidate Recommendation, and considered stable). Returns 10 conformance checks (t01-t10 ⟨8 unchanged words⟩ or any DTCG token export) is structurally correct —: $type/$value/$description present, structured colors (colorSpace + components ⟨54 unchanged words⟩  just its token file), use designesy_score. Executable —: fetches the URL or parses the raw JSON ⟨8 unchanged words⟩  needed. Returns JSON: { checks[{id (t01–t10t01-t10), name, status (PASS/FAIL/WARN), detail}] ⟨22 unchanged words⟩