- Cleanor Tools
Cleanor Tools
Cleanor MCP
Zero-auth, hosted Model Context Protocol server for AI builders.
Two things no popular MCP does well, in one server:
- Actually optimize the images your AI just generated or a user dropped in (WebP / AVIF / JPEG, optional resize) — returns the smaller bytes plus before/after sizes.
- Hand back real, cited Cleanor Labs research data — device storage capacity, next-gen image-format savings, and the "HEIC conversion tax".
No API key. No signup. Every response links its source on cleanor.app.
- Endpoint (Streamable HTTP):
https://mcp.cleanor.app/mcp - Official MCP Registry:
app.cleanor/cleanor - Homepage: https://mcp.cleanor.app
Tools
| Tool | What it does |
|---|---|
optimize_image | Fetch an image URL → re-encode smaller (WebP/AVIF/JPEG), optional resize. Returns the optimized image + before/after bytes. |
storage_capacity | How many photos / minutes of video fit in a given GB tier, corrected for real OS/filesystem overhead. Backed by the photo-storage-capacity study. |
image_format_savings | How much smaller WebP / AVIF / JPEG XL are than JPEG at matched quality, plus the HEIC → JPG/PNG conversion tax. From Cleanor's controlled benchmark. |
qr_code | Text or URL → a crisp, dependency-free SVG QR code you can paste anywhere. |
hash | SHA-1/256/384/512 hex digest of text. |
hmac | Keyed HMAC signature (SHA family), hex or Base64. |
uuid | UUID v4 (random) or v7 (time-sortable). |
base64 | Base64 encode/decode, UTF-8 + URL-safe. |
json_format | Validate and pretty-print or minify JSON, optional key sort. |
jwt_decode | Decode a JWT header + payload (no verification). |
regex_test | Test a regex and return every match and captured group. |
cron_describe | Explain a cron expression and list the next runs. |
unit_convert | Length, mass, data, time, speed, temperature. |
datetime | Current time or a timestamp in any IANA timezone. |
url_parse | Break a URL into scheme, host, path, query, fragment. |
base_convert | Integers between bases 2–36 (BigInt-exact). |
diff | Line-by-line diff of two texts with a change count. |
color | Convert a color between hex, RGB and HSL. |
color_palette | Harmonious palette from one base color. |
placeholder_image | Sized SVG placeholder with a label and colors. |
slugify | Clean, URL-safe slug from a title. |
count | Exact characters, words, lines and UTF-8 bytes. |
The first four are image + cited-data tools; the remaining 18 are deterministic developer utilities for the operations LLMs get wrong. All 22 tools are read-only and safe to expose to autonomous agents.
Connect
Claude Code (CLI):
claude mcp add --transport http cleanor https://mcp.cleanor.app/mcp
Cursor (.cursor/mcp.json), Claude Desktop, VS Code, and most clients:
{
"mcpServers": {
"cleanor": {
"url": "https://mcp.cleanor.app/mcp"
}
}
}
Stdio-only clients can run the server locally via npx — no bridge needed. Published on npm as @cleanor/mcp:
{
"mcpServers": {
"cleanor": {
"command": "npx",
"args": ["-y", "@cleanor/mcp"]
}
}
}
Local image optimization uses sharp (an optional dependency); the data and QR tools work regardless.
About
Built and hosted by Cleanor Labs — free, private, in-browser tools (files never leave your device) plus original research on device storage and image formats. The MCP server is a thin, hosted front door to that toolset and data for AI agents.
- Free browser tools: https://cleanor.app/tools
- Original research / studies: https://cleanor.app/research
License
MIT © Cleanor Labs
Citation
If you use Cleanor MCP or its data, please cite it. Authored by Cleanor Labs, ORCID 0009-0005-4623-961X. The archived, citable version has DOI 10.5281/zenodo.21225551 (concept DOI, always resolves to the latest release).
服务器配置
{
"mcpServers": {
"cleanor": {
"url": "https://mcp.cleanor.app/mcp"
}
}
}