layer-pixel-art
Use when making pixel art with Layer: sprites, retro game characters, pixel icons, pixel tiles, or a sprite sheet and walk cycle. Also when generated pixel art comes back blurry, anti-aliased, or at the wrong grid size, and when packing separate frames into a sheet or cutting an existing sheet into frames. Keywords: pixel art, sprite, sprite sheet, spritesheet, 16-bit, 32x32, walk cycle, animation frames, palette, dithering, pack, split.
- License 许可
- MIT
- 最近更新
- 2026-11-15
- Stars
- ★ 6
- Forks
- 0
01 前言 Front Matter
- name
- layer-pixel-art
- license
- MIT
- description
- Use when making pixel art with Layer: sprites, retro game characters, pixel icons, pixel tiles, or a sprite sheet and walk cycle. Also when generated pixel art comes back blurry, anti-aliased, or at the wrong grid size, and when packing separate frames into a sheet or cutting an existing sheet into frames. Keywords: pixel art, sprite, sprite sheet, spritesheet, 16-bit, 32x32, walk cycle, animation frames, palette, dithering, pack, split.
02 资源文件 1
- 参考文档skill-report.json25.4 KB
03 SKILL.md 正文
Layer Pixel Art
Overview
Pixel art is the generation task most likely to come back subtly wrong: a picture of pixel art, soft-edged and anti-aliased at an arbitrary grid, rather than pixel art. Getting it right is mostly about excluding the defaults.
Layer ships two deterministic sprite-sheet utilities alongside generation. pack_sprite_sheet
assembles separate frame images into one uniform-cell sheet on a transparent background, and
split_sprite_sheet cuts a sheet into its cells in reading order. These are named directly rather
than discovered because they are first-party tools rather than models, with no competing
alternative in the catalogue.
Generation mechanics are in the layer skill. If a sibling skill named here is missing from your
available skills, ask the user to install it (npx skills add layerai/skills --skill <name>);
unattended, proceed from tool schemas and flag the gap.
Name the grid and the palette
Both, every time, as numbers.
- Grid: "32x32 sprite", "16x16 tile". A model given no grid produces something pixel-ish at no particular resolution, which cannot be cleaned up afterwards.
- Palette: a count and a family. "Limited 16-colour palette, muted earth tones" constrains the output in a way "retro colours" does not.
- Era, when the user has one in mind: 8-bit NES-era flat colour, 16-bit SNES-era with shading ramps, or modern high-resolution pixel art. These look nothing alike.
Exclude the defaults, in the prompt
The defects are predictable, so name them as exclusions: no anti-aliasing, no blur, no gradients, no
soft shadows, hard-edged pixels only, one pixel per cell. Where the model declares negative_prompt,
put them there instead. Also state the background: "isolated on transparent background" or "flat
single-colour background", since a generated backdrop has to be cut out and pixel art does not
survive a sloppy cut.
Sprite sheets and walk cycles
Do not ask one generation for a sheet of frames. Models produce a grid of loosely related drawings, and the character changes between cells, which is exactly what a sheet must not do.
The working sequence:
- Generate and approve one frame: the idle or contact pose, isolated on transparent.
- Generate each remaining frame separately, with the approved frame attached as
reference_image, and describe only what changed: "same character, same palette, same 32x32 grid, left leg forward mid-stride, right arm back." - Review the frames as a set and reroll the ones that drifted.
pack_sprite_sheetto assemble them into a uniform-cell sheet. Its input schema is the authority on the argument shape, so read it rather than assuming the parameter name.
Every frame after the first references frame one, not the frame before it. Chaining each frame off the last one compounds drift, so by frame four the knight is a different knight.
When a trained pixel style is in play, attach it as reference_sets and omit base_model_id
entirely, so the server picks a model that can apply it. Use the same set, the same weight, and the
same reference frame on every frame of the cycle.
A four-frame cycle (contact, down, passing, up) reads as motion and is far more reliable than eight. Start there and add frames only if the motion needs them.
Going the other way, split_sprite_sheet cuts a user's existing sheet into individual frames, which
is the way to edit one frame of a sheet without touching the others.
Tiles
A pixel tile that repeats seamlessly is a tiling problem before it is a pixel problem, so filter for
the tileability capability and read layer-textures for how to specify scale and avoid visible
repetition. Everything on this page about grid, palette, and hard edges still applies.
Worked example
"A 32x32 walk cycle for our knight sprite."
- Generate the contact pose.
list_base_modelswithfilter.use_case: "text_to_image", take the first result, then prompt: "32x32 pixel art sprite of an armoured knight, side view, contact pose with left foot forward, limited 16-colour palette, 16-bit era shading, hard-edged pixels, no anti-aliasing, no blur, no gradients, isolated on transparent background." - Approve one. This frame is now the reference for everything else.
- Three more runs, one per remaining frame, each attaching the approved frame as
reference_imageand naming only the pose change. - Check the four together for drift in palette, height, and armour detail. Reroll any that moved.
pack_sprite_sheetwith the fourfile_idvalues, which returns the sheet on a transparent background.
Common mistakes
- Omitting the grid size, then trying to downsample a soft image into pixels.
- Asking one generation for a whole sprite sheet.
- Forgetting the anti-aliasing and blur exclusions, which are the defaults being excluded.
- Generating frames independently with no reference frame, so the character drifts.
- Generating an eight-frame cycle before a four-frame one works.
- Accepting a generated background instead of asking for transparent.
- Packing a sheet by hand when
pack_sprite_sheetproduces uniform cells deterministically.
04 同仓库更多 Skills
serpmaven
Scrape Google search results (SERP) to JSON with SERP Maven (serpmaven) - a real Chrome browser on a residential proxy with sticky sessions. Run it as a CLI, a Python library, a local HTTP search API (`serve`), an MCP server or a Docker image; set the proxy login, workers, pages and cost tracking. Use when the user wants Google results, rank tracking, a self-hosted SERP API, or live Google search for an AI agent.
revnu-landing-page-audit
Find out why a landing page isn't turning visitors into signups, demos or sales. Audits one page against what the ad, post or search that sends people there promised, then checks the first screen, proof, call to action, friction and mobile. Returns at most 7 prioritized fixes, plus what is fine and what could not be checked; rewrites specific lines and fixes the page in the repo when asked. Never promises a conversion lift. Works without a Revnu account. Use when the user says or asks: "audit my landing page", "why isn't this page converting", "review my homepage", "my ads get clicks but no signups", "people bounce from the pricing page", "CRO review", "improve conversion rate", "is my hero section clear", "roast my landing page", "rewrite my headline", "what's wrong with this page", "landing page feedback", "make the signup page convert better". Not for SEO or AI-search audits, page speed engineering on its own, or building a new page from scratch.
revnu-growth-context
Write down once who buys, what they're stuck on, what it costs, what proof you have and what you've already tried, so every growth task starts from facts instead of a questionnaire. Creates or updates .agents/revnu-growth.md from the repo, the live site and the user, and imports an existing product-marketing.md. Use when the user says "set up our marketing context", "who is our ideal customer", "write our ICP", "describe the product for marketing", "positioning doc", "what's our brand voice", "remember this about our customers", "log what we tried", "what have we learned from marketing", "we changed pricing, update the context", "update our growth notes", "summarize the business for the agent", "import my product-marketing.md", "start a growth doc", or before a first marketing, SEO, outreach or launch task when no context file exists. Works without a Revnu account.
revnu-get-listed
Get onto the third-party pages AI answers and search cite when buyers ask for the best tool in your category: "best X" roundups, directories, review sites, GitHub awesome lists, community threads and comparison pages. Finds which ones are cited, ranks them by fit and effort, and drafts each submission, PR or pitch by the target's own rules. Use when the user says "get us listed", "which directories should we submit to", "how do we get into best-of lists", "why does ChatGPT recommend competitors and not us", "get mentioned in AI answers", "submit to awesome lists", "get on G2", "where are our competitors listed", "pitch listicle authors", "find backlink opportunities", "launch directories", "get into roundups", "Product Hunt alternatives", "where should we be listed", "get cited by Perplexity". Never submits, pays or posts without a yes. Works without a Revnu account.
revnu-ai-readiness-audit
Audit whether your website gives ChatGPT, Perplexity, Claude, Gemini and Google's AI Overviews something worth quoting, and fix the pages that don't. Use when the user says or asks "is my site ready for AI search", "audit my site for AI answers", "AEO audit", "GEO audit", "why doesn't ChatGPT quote our pricing", "make our pages citable by AI", "do I need llms.txt", "does schema help with ChatGPT", "check our structured data", "is our content visible without JavaScript", "AI SEO audit", "optimize our homepage for AI answers", "LLM SEO check", "why does AI get our product wrong", "fix our site for AI Overviews". Checks fetchability and indexability, answers in the raw HTML, entity facts, schema (including JS-injected markup), headings, dates and llms.txt, then gives at most 10 prioritised fixes and edits the repo when asked. Works without a Revnu account.
proxy-troubleshooting
Diagnose residential proxy failures, with NodeMaven's gateway as the reference - 407 with correct credentials, 406, 410, 500, a connection that hangs, ERR_TUNNEL_CONNECTION_FAILED in a browser, Playwright or Puppeteer ignoring proxy credentials, a sticky session that changes IP, geo-targeting that is silently ignored, passwords with special characters, and unexpected traffic from headless Chrome. Use when a proxy connection fails, returns the wrong location, or costs more traffic than expected.