This MCP server routes a single prompt to generate developer image assets. It produces app icons, favicons, OG images, logos, splash screens, SVG, and platform bundles by routing across 30+ image models. The workflow is designed to work from the CLI or an AI assistant.
π οΈ Key Features
Generate app icons
Generate favicons
Generate OG images
Generate logos
Create splash screens, SVG, and platform bundles
Routes across 30+ image models
π Use Cases
Produce favicon bundles for IDEs/editors
Generate social/preview images (OG)
Create icon/logo assets from one brief prompt
Use from the CLI or via an AI assistant
β‘ Developer Benefits
Single-prompt to multiple asset types
Supports βzero API keyβ routing (as stated in the excerpt)
Works with Claude Code, Cursor, Windsurf, and VS Code
β οΈ Limitations
Supported output types and routing targets are specific to the serverβs listed asset bundle and model routes
prompt-to-asset demo β one prompt becomes a full favicon bundle in 20 seconds. Works with Claude Code, Cursor, Windsurf, VS Code. Zero API key required.
promptβtoβasset
Free image generation for developers, from the CLI or your AI assistant.
Turn one brief into app icons, favicons, OG images, logos, splash screens, SVG, and platform bundles using the best zero-cost routes first. Built for Cloudflare Workers AI, NVIDIA NIM, HF Inference, Stable Horde, Pollinations, free trials, and offline SVG.
Pick one. Run it. You're done. The recommended path is first.
Runtime: Node β₯ 20.11 (24 recommended). macOS, Linux, Windows (WSL2 for --fix native deps).
π΅ AI assistant Β· recommended
Click an install button above. Then paste one of these into chat:
text
β’ Make a transparent logo for Forge, a dev-tools brand. Flat vector, warm orange.
β’ Make a favicon for my app, dark-mode aware.
β’ Fan this master.png out to iOS + Android + PWA.
β’ Ingest this screenshot as a mark, vectorize, export everything.
Works in Cursor, Claude Code, VS Code, Windsurf, Codex, Gemini CLI. Zero terminal typing.
π’ Zero key, zero install
One curl, one npx. Offline fan-out to every platform.
Best CLI experience: free tokens first, paid keys optional.
bash
npm i -g prompt-to-asset
p2a doctor # check env
p2a doctor --fix # auto-install deps
p2a pick # interactive
Start with Cloudflare Workers AI, NVIDIA NIM, HF Inference, Stable Horde, and free trials. Paid providers remain optional fallbacks.
Free paths at a glance
You don't need a paid API key. Ranked best-first:
Route
Gets you
Signup
inline_svg via AI assistant
Logos, favicons, icon packs β instant
None
Cloudflare Workers AI
Flux-1-Schnell + SDXL, 10k neurons/day
Free token + account ID
NVIDIA NIM
Flux.1-dev, Flux.2-klein, SDXL, SANA
Free token, no card
HF Inference
SDXL, SD3, Flux dev/schnell
Free read token
Stable Horde
SDXL, Flux on community GPUs
Anonymous queue
Pollinations (HTTP GET)
Last-resort raster, RGB
None
Google AI Studio (paste-only UI)
Nano Banana / Nano Banana Pro
Google account
Details + quotas: Free paths beyond Pollinations. Run p2a doctor or ask your assistant for asset_doctor() to see what's live right now.
TIP
Stuck? Click Install in Cursor or Install in VS Code above, restart the editor, and say: "make a favicon for my app, dark-mode aware."
Highlights
Free CLI-first generation β p2a doctor, p2a pick, and the MCP tools rank free API routes before paid models, so developers can generate inside the CLI without enabling billing.
Three execution modes β inline_svg (host LLM authors SVG), external_prompt_only (paste into any web UI), api (server calls a free or paid provider). Pick what fits. All three can finish on $0.
60+ models, free-first routing β Cloudflare Workers AI, NVIDIA NIM, HF, Horde, Pollinations, free trials, plus optional paid OpenAI / Ideogram / Recraft / BFL / Gemini. Each rule cites a research source.
Refuses to do the wrong thing β the Never column. No wordmarks past 3 words through a diffusion sampler. No transparent PNG through Imagen. No negative_prompt on Flux.
You just talk to your assistant. Example from a new chat:
Make me a transparent logo for a developer-tools company called Forge. Flat vector, two-tone warm orange on neutral.
Behind the scenes:
asset_doctor() β check what modes and providers are live.
asset_init_brand({ app_name: "Forge", palette: ["#EA580C", "#F5F5F4"] }) if no brand.json exists.
asset_enhance_prompt({ brief }) returns an AssetSpec: classification, rewritten prompt, modes_available[], optional svg_brief, optional paste_targets, and a routing_trace pointing at the research file that backed the decision (plus never_models β why Imagen or DALLΒ·E got rejected).
Assistant offers you inline_svg / external_prompt_only / api.
If inline_svg: it writes <svg> inline and calls asset_save_inline_svg β writes master.svg, favicon.ico, apple-touch, AppIconSet, PWA bundle to disk.
If external_prompt_only: assistant shows the refined prompt and the best paste target (free first). You generate, save, then say "ingest this file" β asset_ingest_external.
If api: assistant calls the routed provider. Server mattes, vectorizes, exports, validates.
Follow-up: "also fan this out for iOS and Android" β asset_export_bundle with the saved master.
Zero CLI typing. The CLI is still first-class for CI, shell scripts, and non-MCP environments β both surfaces hit the same core.
Why this exists
Two facts shape everything here.
Producing production-grade software assets is a routing and post-processing problem, not a prompt-engineering problem.
Imagen 3/4 and Gemini Flash Image can't produce real RGBA PNGs β their VAE is RGB-only, so asking for a transparent background renders the grey-and-white checkerboard as pixels. SDXL can't spell past ~8 characters. Only Recraft emits native SVG. Flux errors on negative_prompt. None of that is visible in the model UI. All of it silently breaks one-shot "prompt β asset" tools.
You may not have an image-model API key. The plugin works anyway.
Every one of the three modes can finish on $0.
The three modes
flowchart LR
A["one-line brief"] --> B["asset_enhance_prompt"]
B --> C{"mode"}
C -->|inline_svg| D["Host LLM emits SVG inline<br/>β asset_save_inline_svg"]
C -->|external_prompt_only| E["Paste into web UI<br/>β asset_ingest_external"]
C -->|api| F["Server calls routed provider"]
D --> G["matte Β· vectorize Β· validate"]
E --> G
F --> G
G --> H["AssetBundle<br/>ios Β· android Β· pwa Β· favicon Β· visionos Β· flutter"]
Server returns the dialect-correct prompt plus a ranked list of paste targets, free paths first: Pollinations, HF Inference, Stable Horde, Google AI Studio, Ideogram, Recraft, Midjourney, fal.ai, BFL, ChatGPT, Firefly, Krea. Generate elsewhere, save locally, call asset_ingest_external.
Anything β best for illustrations, heroes, text-heavy logos
api
Optional
Server calls the provider directly. Works zero-key via Pollinations / Horde / HF, or with paid keys. Route β generate β matte β vectorize β export β validate β content-addressed bundle.
Automation, CI, no rate-limit tolerance
The host LLM picks the mode, or you do. The server surfaces modes_available so the assistant offers them to you. Free paths first β always.
Never a diffusion sampler. Mark + SVG typography composite.
β
β
Native SVG
Recraft V4 (V3 for brand-style pipelines)
inline_svg (host LLM authors SVG)
Everyone else
Photoreal hero
Free CLI route first: Cloudflare / NVIDIA NIM / HF
Flux Pro / gpt-image-1.5 / Gemini with paid keys
DALLΒ·E 3, Imagen 4 for new work
Iterate an existing mark
flux-kontext-pro (edit-only)
Pollinations Kontext (free)
β
Zero-cost everything
inline_svg β Cloudflare / NVIDIA NIM / HF / Horde
Pollinations β paste-only
β
The Never column matters. It's why prompt-to-asset refuses to render wordmarks past 3 words in any diffusion sampler, and why asking for a transparent PNG never goes to Imagen.
Free paths beyond Pollinations
Option
How
Best at
Catch
Cloudflare Workers AI
Free API token + account ID
Flux-1-Schnell, SDXL, DreamShaper
10k neurons/day cap (~900 Flux-Schnell or 5k SDXL-Lightning)
NVIDIA NIM
Free API token, no card
Flux.1-dev, Flux.2-klein, SDXL, SANA
1k requests/month; strong free CLI route
HF Inference
Free read token
SDXL, SD3, Flux dev + schnell
Rate-limited, cold-start latency
Stable Horde
Anonymous kudos queue
SDXL, Flux community GPUs
Minutes of queue on the free lane
Pollinations.ai
curl β HTTP GET. No signup.
Throwaway raster
Last resort: silent model swaps, downsizing, RGB only
Google AI Studio (UI)
Free interactive web UI at aistudio.google.com
Nano Banana / Nano Banana Pro
No free API β paste-only; download PNG, call asset_ingest_external
Local ComfyUI
Community comfyui-mcp adapter
Full fidelity, no caps
You bring the GPU
inline_svg
Host LLM emits <svg> in chat
Logos, favicons, simple icons
β€40 paths; simple geometry
external_prompt_only
Paste into any web UI
Whatever that UI gives you
Manual save, then asset_ingest_external
Google Gemini / Imagen β verified pricing (Apr 2026)
Verified 2026-04-26 against public Gemini API pricing:
Imagen 4 image output is paid API. The pricing table lists Free Tier as "Not available" and prices Fast / Standard / Ultra at $0.02 / $0.04 / $0.06 per image.
Nano Banana family is paid API.gemini-3.1-flash-image-preview (Nano Banana 2), gemini-3-pro-image-preview (Nano Banana Pro), and gemini-2.5-flash-image (original Nano Banana) list image output only in the paid tier. Billing must be enabled on the GCP project to use these via API.
Free for text, multimodal understanding, and embeddings. The Gemini text-out models still have Free of charge input + output on the free tier.
Free interactive image generation is also via the AI Studio web UI at aistudio.google.com. Community-observed limit 500β1,000 images/day, dynamic. Use external_prompt_only + asset_ingest_external if you want Nano Banana quality without billing.
Free image generation via the Gemini consumer app at gemini.google.com: Basic 20/day, AI Plus 50/day, AI Pro 100/day, Ultra 1,000/day.
Paid API pricing (per image, standard): Nano Banana (gemini-2.5-flash-image) $0.039; Nano Banana 2 Flash (gemini-3.1-flash-image-preview) $0.045/0.5K, $0.067/1K, $0.101/2K, $0.151/4K; Nano Banana Pro (gemini-3-pro-image-preview) $0.134/1K-2K, $0.24/4K (+ $0.0011 per input image); Imagen 4 Fast $0.02, Standard $0.04, Ultra $0.06. Batch API is 50% off.
Run p2a doctor (or ask your assistant for asset_doctor()) to see what's live in your environment right now.
Install
Every command works via npx β no install required.
bash
# Zero install
npx prompt-to-asset doctor # what's live in this shell right now
npx prompt-to-asset doctor --fix # auto-install native deps (brew / cargo / scoop; never sudo)
npx prompt-to-asset pick # interactive route picker
npx prompt-to-asset init --register # scaffold brand.json + register in .cursor / .vscode / .windsurf# Or global for daily use
npm i -g prompt-to-asset
p2a doctor
# Or per-project for CI
npm i -D prompt-to-asset
Register with your AI assistant
Claude Code
bash
claude mcp add prompt-to-asset -- p2a
Smithery (universal)
bash
npx -y @smithery/cli install prompt-to-asset --client claude
Paste-only surfaces:midjourney-v6, midjourney-v7, firefly-3, krea-image-1. Calling asset_generate_* with mode: "api" against a paste-only primary soft-falls-back to the first API-reachable model in the chain and surfaces a warning. If the whole chain is paste-only, you get an ExternalPromptPlan rather than an error.
favicon.ico (16/32/48 multi-res), icon.svg with prefers-color-scheme dark support, apple-touch-icon.png 180Γ180 opaque, 192/512/512-maskable, manifest.webmanifest, <link> snippet for your <head>.
Content-addressed storage: assets/<hash[0:2]>/<hash>/<variant>.<ext>. The MCP server is synchronous and stateless. prompt_hash and params_hash in every AssetBundle are designed to drop straight into a BullMQ / SQS / Cloudflare Queues jobId for a hosted pipeline. Reference design: docs/research/18-asset-pipeline-tools/18e-production-pipeline-architecture.md.
Design thesis
You own the API keys. The LLM owns everything else.
The only thing that happens in a terminal is installing the package and putting keys in .env. Secrets shouldn't pass through chat. Every other verb β doctor checks, model inspection, platform fan-out, brand scaffolding, sprite sheets, 9-slice configs β is an MCP tool the assistant calls when you ask in natural language.
This tool handles API keys for up to 15 providers. Non-negotiables:
Keys live in env vars only. Never written to disk, never logged, never echoed in MCP responses. Provider error bodies go through redact() (packages/mcp-server/src/security/redact.ts) before reaching the host LLM.
Path access is allow-listed.image_path / output_dir / existing_mark_svg resolve through symlinks and reject anything escaping project cwd + configured output dir + cache dir + OS tempdir. Widen with P2A_ALLOWED_PATHS=/path1:/path2.
SVG is XSS-sanitized before any write.<script>, <foreignObject>, on*= handlers, javascript: URIs, external <image href> / <use href>, CSS @import over the network β all rejected. The check runs unconditionally; SVGO is not required.
Cost guardrail. Set P2A_MAX_SPEND_USD_PER_RUN=5.00 to cap any single tool call. Pre-flight estimate refuses to call if over. Free-tier routes are always $0.
Data integrity at boot.assertDataIntegrityAtBoot() runs on start. If a routing rule points at a model id not in the registry, the server refuses to boot with a clear error. Check in CI with p2a doctor --data.
No telemetry. No remote calls unless the routed provider explicitly requires one.
Every routing rule, dialect switch, safe-zone size, and text ceiling that's implemented is backed by a file under docs/research/. asset_enhance_prompt returns a routing_trace.research_sources array on every call. The angle β code pointer map, plus an honest ledger of what's wired and what's deferred, lives in docs/RESEARCH_MAP.md.
Development
bash
git clone https://github.com/MohamedAbdallah-14/prompt-to-asset.git
cd prompt-to-asset
npm install
npm run build
npm run typecheck
npm run lint
npm test# vitest watch
npm run test:run # vitest run (CI)
npm run smoke # list tools via MCP stdio + correctness assertions
npm run sync# regenerate IDE mirrors from SSOTs
npm run verify # byte-verify mirrors match SSOTs
SSOTs live in skills/*/SKILL.md, rules/*.md, .claude-plugin/, and data/*.json. Don't edit .cursor/, .claude/, .windsurf/ directly β they're regenerated by scripts/sync-mirrors.sh and CI byte-verifies them.