Profile note: this skill's tools are search_sites, get_site_details, get_site_elements, list_categories, get_annual_categories, get_annual_nominees, get_annual_nominee, get_collections, search_directory, new_winners, watch_site, capture_live_site, analyze_page_structure, record_site_motion and retained create_design_brief. Examples below use 1.x names (default full). On AWWWARDS_MCP_PROFILE=v2, use search_references {request:{kind:"sites", ...}}, get_reference {request:{kind:"site_details", slug}} or get_reference {request:{kind:"site_details", slugs}} for a batch of 1-10, get_reference {request:{kind:"site_elements", slug}}, browse_awards {request:{kind:"filters"}}, browse_awards {request:{kind:"annual_categories"}} (also kind:"annual_nominees" with category, and kind:"annual_nominee" with id or category + slug), search_references {request:{kind:"collections"}} (or get_reference {request:{kind:"collection", slug}} for one collection), search_references {request:{kind:"directory", ...}}, monitor_awards {request:{action:"poll_winners"}}, monitor_awards {request:{action:"add_watch", kind, pattern}}, monitor_awards {request:{action:"list_watches"}}, monitor_awards {request:{action:"remove_watch", kind, pattern}}, capture_page {request:{kind:"screenshot", url}}, inspect_page {request:{kind:"structure", url}} and capture_page {request:{kind:"recording", url}}; create_design_brief stays flat. Full mapping: v2-routing.md.

When to use this skill

First time on this machine? If ~/.awwwards-mcp/preferences.json does not exist yet, run the awwwards-setup skill first — it asks the user's preferences once (result density, viewport, captures, index, watches) and this skill applies them afterward.
During site builds — before writing any UI code, gather references and state a design direction.
Standalone research — the user wants inspiration, trends, or examples ("show me dark 3D portfolio sites").
Index & capture ops — building the local search index, or taking fresh screenshots of live URLs.

The inspiration loop

Run this loop before building anything visual:

Restate the goal as concrete attributes. Turn the user's request into mood, color, technology, and industry terms. "Make it feel premium" becomes e.g. "dark, elegant, WebGL, agency portfolio".
Ground your vocabulary. If unsure which filters exist, call list_categories first — it returns every color hex, award filter (sotd, soty, nominees, winner categories, …) and tag/technology slug you can search by. For browsable discovery, get_collections lists curated/member collections (pass a returned ref as slug for one collection's cards), and search_directory lists agencies and studios with optional country/category/budget filters. For annual "of the Year" questions, the get_annual_* tools cover the six official annual categories (see Award discovery below).
Search. Call search_sites with 1–3 filters (e.g. { color: "#404040", tags: ["3d", "portfolio"] }). Judge the results from the inline screenshots, not just titles. Shortlist 2–3 candidates. Free-text queries are porter-stem + prefix-matched and BM25-ranked ("magazines" finds Magazine-tagged sites, best matches first); multi-word queries keep AND semantics — both tokens must hit the same site. 3b. Techniques registry: _memory/techniques.json catalogs researched how-tos per domain (video-understanding, motion-detection, ui-structure, micro-interactions, images). Consult it when a step needs a method — it grows via the flywheel.
Synthesize the direction. Once you have 2–5 shortlisted slugs, call create_design_brief with those slugs and the user's non-blank project goal. Optionally include audience and concrete constraints. Use the returned shared signals, deliberate outliers, build constraints, evidence, limitations, and Interaction Plan to define an original direction — do not copy any single reference. Apply only its evidence-backed cursor/hover, scroll/pin/scrub, entrance/reveal, easing, and duration cues. Treat empty or null interaction fields as missing evidence: do not invent effects, timings, or scroll behavior. Preserve the plan's prefers-reduced-motion, content-access, and keyboard-access guardrail. Set includeStructure: true only when optional page-band evidence is needed; it may run browser analysis for sources with usable live URLs. Cached Motion DNA is included when available, but the brief never records motion automatically.
Live reference URL named? Capture it full-page first. (and later capture your own build the same way — compare both against each other) If the user points at a specific live site (e.g. "recreate cerebrium.ai"), call capture_live_site on that URL before anything else and design from the full-page PNG — every section, top to bottom. If analyze_page_structure reports scrollRoot (an inner scroller), the capture returns mode:"anchors" instead — six ordered viewport PNGs down that container as JSON paths (no inline image); read the six files. Cached Awwwards screenshots are hero-only crops (~880×660) and hide everything below the fold: the sections that make a site's structure distinctive (pricing, feature layouts, contrast breaks, footer) were never visible in them. For mobile-excellence references pass viewport: "mobile" — and capture BOTH viewports when the desktop and mobile designs diverge.
Get the design DNA as follow-up evidence. Call get_site_details on the top pick when the brief reports missing or weak detail evidence, or when a specific source needs deeper inspection. If it reports a layout-drift error, fall back to judging the shortlisted screenshots, get_site_elements (which uses a different parser), or a capture_live_site of the site's URL for a first-hand capture. For a shortlist, get_site_details { slugs } gives bounded enrichment: up to 10 sites in one call, compact details, no images.
Study MOTION before building (element videos, not posters). Call get_site_elements on shortlisted sites — it returns per-element video URLs (preloader, page transition, case study, about…). Download those videos and tile them at 1–2 fps BEFORE writing any animation code: curl -o e.mp4 <video-url> && ffmpeg -i e.mp4 -vf "fps=2,scale=480:-1,tile=6x4" -frames:v 1 tile-e.jpg, then Read the tile. (The full capture→review→build procedure — what to record, what to skip, motion inventories, build verification — is the awwwards-motion-study skill.) Element posters are single frames — they show layout, not motion; a 3D carousel looks like floating static cards in a poster (this exact mis-build happened). One tile per element shows you the whole animation arc (easing, overlap, entrance order). If a marquee animation needs finer study, re-tile that video at higher fps. The motion IS the design: every reference animation you ship must trace to frames you actually studied, and reference videos pass through showcase/ref-motion/ as the design's evidence trail.
State the design direction before writing code. In prose: palette (hexes from the references), type mood, layout patterns, page architecture (one page vs separate horizontal sections — judge this from the reference passages you studied in steps 4–6, not habit), and tech choices, each traceable to a reference. Then build. For a site's own marquee frontend skills, prefer proven patterns (GSAP ScrollTrigger for horizontal scroll chapters, Lenis for smooth scroll) over hand-rolled scroll math.
Verify structure, then polish. After building, capture your own build with capture_live_site (on its file:// or served URL) and run analyze_page_structure on BOTH the reference and the build. Compare band maps section by section (count, order, backgrounds, heights). Match the reference's viewport when comparing: for mobile-excellence references pass viewport: "mobile", and capture BOTH viewports when the design diverges. Fix distribution mismatches first — a section that is 3× the reference's height is a structural bug no amount of pixel polish fixes. Match the reference's band structure, never just its total height.
Close the flywheel: in a checkout of the awwwards-skills repository (the installed skill copy does not include this dev script), end every loop pass by recording what the verification caught - node scripts/skill-memory.mjs record --skill awwwards-inspiration --phase <phase> --symptom "..." --rule "..." [--evidence path]. "Nothing new learned" is a recorded negative. Then distill folds rules seen ≥2× into your installed copy's managed section; recall prints them at loop start. When promoting a confirmed rule into the shipped copy (human PR), promote --skill awwwards-inspiration prints it with machine-local evidence paths stripped.

Award discovery

Beyond the daily listing sweep, Awwwards exposes an annual-awards layer and finer winner cuts:

Annual awards (official API) — for "Site of the Year"-style questions, start with get_annual_categories, then get_annual_nominees (per category, hydrated at 1 req/s) and get_annual_nominee for full structured nominee detail. These cover the six official annual categories.
Daily / monthly / yearly / nominee cuts — search_sites with award: "sotd" | "sotm" | "soty" | "nominees" (plus developer / honorable).
Annual winner categories — search_sites with award: "ecommerce" | "portfolio" | "product" | "typography" | "businessServices" | "noCode".
Unsure of a slug? Call list_categories first — it returns every award filter with its label plus the current technology slugs.

Anti-patterns

Vague single-word searches ("modern", "nice") — use concrete color/tag/ technology/award filters instead.
Skipping to code without stating a direction — the references are worthless if nothing is derived from them.
Dumping raw tool output at the user — curate: show the shortlist, the chosen direction, and why.
Designing from a thumbnail or element poster — thumbnails are hero-only crops, posters are single video frames, and neither shows real structure OR motion. When the reference URL is known, capture it full-page (step 4); when an element has motion, tile its video (step 6) — never animate from posters.
One filmstrip for everything — a single N×N tiling of the whole recording makes late tiles tiny and animation arcs illegible. Tile per element/per passage at 1–2 fps instead; re-tile finer passages at higher fps when easing or overlap matters.
Padding empty bands to match total height — if your build's total height matches the reference but a spacer/background band is far taller than the reference's equivalent, the height was stolen from real content sections. Compare band maps, not totals. Spacer and decorative elements are measured against the reference's equivalent band — never invented to absorb height.

Tool reference

Tool	Key params	Returns	Gotchas
`search_sites`	`query` (free text vs titles/tags), `color` (hex like `#404040`), `tags` (array of slugs), `technology` (slug), `award` (`sotd`\|`sotm`\|`soty`\|`nominees`\|`developer`\|`honorable`\|`ecommerce`\|`portfolio`\|`product`\|`typography`\|`businessServices`\|`noCode`), `count` (1–12, default 6), `page` (default 1), `responseMode` (`"full"` default or `"compact"`)	Full: text site cards + inline JPEG screenshots for each result. Compact: concise cards + up to two inline previews for the requested page.	Awwwards applies only one URL filter — priority color > award > technology > first tag; the rest are checked client-side (common technology spellings like `nextjs`/`three.js` are canonicalized to the live slugs). Winner-category filters verify by URL only and always re-scrape; nominee filters verify via the persisted nominee flag and are served from cache when fresh. Color searches always scrape live (never cached). Deep pagination is unavailable by design (robots.txt). On live-request failure, stale cache is served when present. Compact is opt-in for broad exploration; use default full results to judge screenshots and details/motion tools for shortlisted sites.
`get_site_details`	`slug` (from `search_sites`, e.g. `l-i-s-a`)	Title, live URL, awards, color palette, technologies, design elements, description, full-size screenshot URL; inline screenshot when available	Cached 7 days; a parse that comes back all-empty is an error, not a quiet empty result.
`get_site_elements`	`slug`	Numbered element list (image or video, with video URLs) + up to 8 inline poster JPEGs	Videos are mp4 URLs (posters only are shown inline). Elements feed from the same fetch as `get_site_details`.
`list_categories`	none	JSON: award filters (with labels), every color hex and filter/tag slug, technology slugs, plus usage guidance	Cached 30 days. Call this whenever filter vocabulary is uncertain — awards and technology slugs included.
`capture_live_site`	`url` (absolute URL), `viewport` (`"desktop"` default or `"mobile"`), `scrollCapture` (`"auto"` default, `"anchors"` or `"viewport"`)	Document scroller: full-page PNG saved to disk + inline image. Inner scroller (`scrollRoot`): six ordered viewport shots as JSON `anchors[].scrollTop`/`file`, no inline image; `scrollCapture:"viewport"` gives one explicit viewport shot	Requires the optional playwright dependency (`npm install -g awwwards-mcp playwright ffmpeg-static && npm exec --global -- playwright install chromium`). After installing, switch the MCP registration from `npx -y awwwards-mcp` to the installed `awwwards-mcp` command and restart, so captures resolve Playwright from the server's own install.
`analyze_page_structure`	`url` (absolute URL or `file://` path), `maxBands` (cap on returned bands, default 40)	JSON: `title`, `totalHeight`, and an ordered band map (`index`, `tag`, `label`, `background`, `offsetTop`, `height`, `textStart` (first ~60 chars of the band's text) per band)	Requires playwright. Run it on BOTH the reference and your build (step 8) and compare band maps — count, order, backgrounds, heights — never just total height.
`record_site_motion`	`url` (absolute URL), `frames` (filmstrip tile count, 4–36, default 16 → a 4x4 grid)	Inline filmstrip JPEG of a motion-through pass (preloader dwell, slow scroll, hover/cursor interactions) + the saved .webm path as text	Requires playwright + ffmpeg-static. Runs a ~30 s scripted pass — heavier than a capture, use when motion matters (step on from static captures).
`new_winners`	`award` (`sotd` default, or `developer`/`honorable`)	Today's freshly-crowned winners since the last poll, as site cards; first-ever call seeds a baseline and dumps the listing instead	UTC-day granular. Each first-seen winner's Elements section is auto-backfilled into the `source:"site"` element corpus — new winners are searchable by elements immediately without gallery indexing. Longitudinal monitoring: call periodically during a session to catch new SOTDs as they land.
`watch_site`	`action` (`add`\|`list`\|`remove`), `kind` (`studio`\|`tag`\|`url`), `pattern`, optional `award`, `note`	list: matching sites per watch + `NEW since last check` deltas; add/remove confirm	Never fetches — evaluates against the freshest cached listing, so run `new_winners` or `search_sites` first (the first `list` only seeds its baseline; a second `list` after rows load reports deltas). url watches take the site slug (e.g. `l-i-s-a`); studio watches are case-insensitive substrings.

All image results arrive as MCP image content blocks — look at them, don't just read the text blocks.

On AWWWARDS_MCP_PROFILE=v2 the watch_site row becomes monitor_awards: add is {request:{action:"add_watch", kind, pattern}}, list is {request:{action:"list_watches"}}, and remove is {request:{action:"remove_watch", kind, pattern}} — kind and pattern are required for both mutations (v2-routing.md).

Staying current

For research that should reflect what's winning now (trend questions, "latest" anything), open the session with new_winners — it reports today's freshly-crowned SOTD/Developer/Honorable winners and quietly folds each winner's elements into the searchable corpus. To track a studio, tag, or specific site over time, add a watch (watch_site, action: "add", kind: studio|tag|url) after your first sweep; every later new_winners / search_sites run refreshes the pool and watch_site list reports what's new per watch. Watches never fetch on their own — the sweep feeds them.

Index & capture ops

Local index. search_sites works out of the box but unindexed depth is limited by polite live scraping (~31 sites per filter page). Build the index once for searches across thousands of sites:

npx -y -p awwwards-mcp awwwards-index      # from the published package
npm run index                              # from the awwwards-mcp checkout root

Crawls all ~200 tag pages at 1 request/second (~4 minutes) into a SQLite cache at ~/.awwwards-mcp/.
Resumable: interrupt and re-run; completed pages are skipped.
The MCP server re-indexes automatically in the background whenever the index is stale — you rarely need to run this by hand.

Live captures. capture_live_site needs playwright installed once (see table above). Two uses: (1) the user wants a screenshot of a URL that is not an Awwwards site, or a fresher view than the cached thumbnails; and (2) — the more important one — the user names a live site as the design reference for a build: capture it full-page first and derive the structure from that image (see step 4 of the loop).

Motion capture. Static images can't show preloaders, scroll-driven animation, or transitions — and most award-winning sites are built around exactly those. When the reference site has motion, record it: launch Playwright with recordVideo, wait out the preloader (~7 s), scroll slowly to the bottom in small steps so every scroll-triggered animation fires on camera, then extract a filmstrip of frames with ffmpeg (npm i -D ffmpeg-static, one frame every ~3 s) and Read the frames — or call record_site_motion directly: the filmstrip comes back inline and the .webm path as text.

Filmstrip is the fallback, not the default. First try Reading the video file directly — if your model supports video input, watching the scroll-through gives you timing, easing, and transitions the frames can't. If the Read comes back with media omitted / "model does not support video input", extract the filmstrip and Read the frames instead. A ready-made recorder ships in the awwwards-mcp source checkout (github.com/INSANE0777/Awwwards-mcp), not this skills repo, at scripts/record-scrollthrough.mjs — run it from that checkout's root (playwright and ffmpeg-static are devDependencies there). Installed/npx users do not get the script; use record_site_motion.

awwwards-inspiration

Awwwards Inspiration