Verified conversions across 205 color spaces: CSS strings, gamut checks, .cube LUTs
Nothing to run from here
nudg is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
94/100
trust score
11
tools declared
local
runs on your machine
—
downloads a day
What it does
Write, check and compare UI interactions (springs, gestures, states) for web, iOS and Android.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user nudg-mcp '--' npx -y @nudg/core mcp
# 2 — ask your agent for something these tools do
# nudg_guide, nudg_recipe, nudg_validate and 8 more
nudg costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
SEAll 1 server
seukwoo1 server
1 category
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts nudg as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
Claude Codedeclared by this listing
This is a public server: you run it yourself and this marketplace is not in the path. You need Node.js — the current LTS (24) is the safe choice; npx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user nudg-mcp '--' npx -y @nudg/core mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it runs from its publisher’s own package: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it writes
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
What you approve
11 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
The nudg format in one page: skeleton, vocabulary, motion forms, structure rules, and which field to change for requests like "bouncier" or "snappier". Read it before writing your first nudg file.
No parameter schema on file.
Turn a common UI term such as toggle, FAB, sheet, voice call, video chat, or media player into a stable interaction contract: required and optional parts, states, events, continuous inputs, host responsibilities, motion, known format gaps, and a validated starter example. Omit name to list…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | no | Recipe name or human alias, e.g. switch, toggle, video chat, sheet, or audio player. |
| source | boolean | no | Include the starter .nudg.json source. Default false. |
Check a nudg file. Each problem has a stable code, a JSON pointer path, and a hint that says how to fix it. Pass file, or source to check text before writing it.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| file | string | no | Path to a .nudg.json file, relative to the working directory. |
| rev | string | no | With file: read it as it was at this git revision (HEAD, HEAD~1, a branch) instead of from disk, imports included. |
| source | string | no | The file's JSON text, to check before writing it. Imports resolve relative to path, or the working directory. |
| path | string | no | With source: where the text would live, for resolving imports and labeling. |
Send events at given times without rendering anything, and get active states and property values over time, plus a per-property summary: min, max, final value, and when it came to rest.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| every | number/string | no | Seconds between samples. Default 0.05. |
Draw a motion graph image: each property over time with state bands, events, targets, overshoot (max/min) and settle times. Good for judging feel, e.g. comparing bounce before and after an edit.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| out | string | no | Also save the graph here (.svg or .png). |
Two versions of an interaction under the same events, for a person to judge. Says first whether anyone would notice the difference (imperceptible / subtle / clear, per property, with what changed: overshoot, settle time, position), then shows both: with html, real A/B filmstrips in two rows;…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| a | object | no | The current version: { file }, { file, rev } (at a git revision) or { source, path? }. Not needed with before. |
| b | object | no | The changed version: { file }, { file, rev } or { source, path? }. Not needed with before. |
| labels | array | no | Names for the two versions, default ["A", "B"]. |
| before | string | no | Shorthand with file: compare the file at this git revision (HEAD, HEAD~1, a branch) with the file on disk; a and b are then not needed. Labels default to the revision and "now". |
| file | string | no | With before: the file to compare across revisions. |
| url | string | no | The running app's URL (or a local page). The component is snapshotted from it, styles and all, so no html file is needed: real A/B filmstrips, and the side-by-side page, on the real thing. Needs Chrom |
| selector | string | no | With url: which element to snapshot. Default: the smallest element holding every bound target (a toggle's track with its knob). |
| open | boolean | no | Open the side-by-side page in the person's browser. Default: once per session; later calls rewrite the same page and the open tab shows it after a reload. true forces a new tab, false never opens (NUD |
| html | string | no | The host page with the component's markup, when there is no running app. Gives real A/B filmstrips (needs Chrome) and writes a page the person should open: both versions playing side by side in a loop |
| page | string | no | Where to write that page. Default: next to out as .html, else in the temp folder. |
| at | array | no | Frame times for the filmstrips. |
| frames | integer | no | Evenly spaced frames when at is not given. Default 6. |
| out | string | no | Also save the picture here (.png; .svg for the graph). |
| bundle | string | no | For a SwiftUI or Compose app: write the comparison (both versions compiled, the events, the verdict) to this .compare.json, for NudgCompare to play side by side on the app's own views. Put it in the a |
Draw real frames of the component as one filmstrip image, using the host page's markup and styles (elements with data-nudg names). Needs a local Chrome.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| file | yes | — | |
| html | string | yes | Page with the component's markup and styles. Its scripts don't run. |
| events | no | — | |
| start | no | — | |
| reducedMotion | no | — | |
| at | array | no | Frame times. Default: evenly spaced until at rest. |
| frames | integer | no | Number of evenly spaced frames when at is not given. Default 6. |
| until | no | — | |
| out | string | no | Also save the PNG here. |
Whether the motion is good, not only valid. Plays every transition from rest and checks it against the platforms' motion guidelines and accessibility rules: transitions that take too long, press feedback that arrives late, changes nobody can see, fades that bounce, exits slower than entrances,…
No parameter schema on file.
The TypeScript module nudg generate writes beside a file (x.nudg.json -> x.nudg.ts): a types-only interface with the file's targets, handled events, continuous inputs and state paths, imports resolved. Hosts pass it to the adapters, e.g. useNudg<ToggleContract>(toggle), so misspelled names fail…
No parameter schema on file.
Prepares a counts-only summary of this machine's nudg log (rounds per change, verdicts, tool calls, versions; no file names, paths, URLs or code) and a link to a prefilled GitHub issue the person can read and submit. Sends nothing by itself. Call it only when the person agreed to share, typically…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| note | string | no | What the person said made it hard, in their words. Optional. |
Host truth. Opens the real page (its scripts run) in headless Chrome, performs actions on the element as a finger or a key would, measures it every frame (translate, scale, rotation, opacity, size, and whether it is still there), and says what the page actually showed. With file, judges the page…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| url | string | yes | An http(s) URL of the running app, or a local .html path (served from the working directory). |
| selector | string | no | CSS selector of the element to measure; shadow roots are searched. Default, with file: [data-nudg="<the file's first target>"]. |
| actions | array | yes | What to do, in time order: { "action": "tap", "t": 0.1 }. Actions are …. The page's own handlers run, so a click that navigates away will. |
| file | no | — | |
| on | string | no | Which of the file's targets the actions are for. Default: the first target that handles them (a toggle's track, not its knob). |
| events | no | The file's events to judge against. Default: the actions, as the file's events on on, at the times they actually happened. | |
| start | no | — | |
| reducedMotion | no | — | |
| settle | number/string | no | Seconds to keep measuring after the last action. Default 1.5. |
| warmup | number/string | no | Seconds to let the page's scripts start first. Default 0.5. |
| viewport | object | no | Window size, e.g. a phone's 390×844. |
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under MIT
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install nudg or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
ATrust score 94 / 100
Release 1.5.0 · scanned 30 Oct 2026
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
Threat score100
Client score94
Findings3 threat · 0 capability
Capabilityhigh
Coveragesource
Verificationsource
Surface digest320c9b75286fa2f2c9a02efba49f7d74e560876a94c6f226db81e9ea7f03170c
Score vectorthreat 100 → client 94
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| verification-discount | source | publisher verification (provenance) — cryptographic build provenance ties the artifact to its source | −0 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
Tool surfaceread
Implementation sourceread
Package metadataread
Live transportnot available
Tools were statically extracted from the published source (11 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. high is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| nudg_guide | no tags | |
| nudg_recipe | no tags | |
| nudg_validate | no tags | |
| nudg_simulate | no tags | |
| nudg_plot | state-write | state-write: declares readOnlyHint: false |
| nudg_compare | state-write | state-write: declares readOnlyHint: false |
| nudg_render | state-write | state-write: declares readOnlyHint: false |
| nudg_critique | no tags | |
| nudg_generate | no tags | |
| nudg_share | no tags | |
| nudg_observe | no tags |
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.
Supply chain and provenance
Integrityfirst-seen
Surface digest320c9b75286fa2f2
Scanned2026-10-30
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 1.5.0. Pin a release below and the command asks for exactly that one.
1.5.0
undated
latestNo release note was published with this version.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish nudg can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.