Name: Electron Viewer
Author: acertijosapps

Electron Viewer MCP

Drive Electron apps over CDP: screenshots, clicks, drags, keys, JS in renderer and main, logs.

Grade A Local · stdio No auth needed 16 tools v1.0.0 MIT

No reviews yet by acertijosappsnpm Runs on your machine, nothing hosted

93/100

trust score

tools declared

local

runs on your machine

—

downloads a day

What it does

Drive Electron apps over CDP: screenshots, clicks, drags, keys, JS in renderer and main, logs.

Quickstart

# 1 — add it to Claude Code
claude mcp add --scope user electron-viewer-mcp '--' npx -y @acertijosapps/electron-viewer mcp

# 2 — ask your agent for something these tools do
#   list_windows, launch_app, attach and 13 more

Electron Viewer costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.

Publishernot claimed

acertijosapps2 servers

1 category

All 2 servers

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 Electron Viewer 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 electron-viewer-mcp '--' npx -y @acertijosapps/electron-viewer mcp

Reconnect, 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 asks of you

Nothing is required, but it can take ELECTRON_VIEWER_URL (Viewer to show actions live (default http://localhost:8094); without it, CDP is used directly), ELECTRON_VIEWER_PROJECT (Electron project folder that launch_app starts with `electron .` (default: the current folder)) if you want to set them.

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

16 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.

Only tools that write or reach the network

Read from its source code. A server that builds its tool list when it starts can list more tools than this.

Lists the running Electron apps (found on CDP ports 9222–9231, launched or attached) and their windows.

No parameter schema on file.

Starts the Electron project with electron . (its own node_modules/electron) on free debugging ports,

Parameter	Type	Required	Notes
project	string	no	Project folder (default: the folder the viewer or MCP was started in)

Attaches to an Electron app already running with --remote-debugging-port=<port> (and optionally

Parameter	Type	Required	Notes
port	integer	yes	CDP port (--remote-debugging-port)
inspect_port	integer	no	Main process inspector port (--inspect)

Captures the window content (PNG). For click, drag, hover and scroll use normalized coordinates:

Parameter	Type	Required	Notes
window	string	no	Window: target id from list_windows or part of its title. Optional when only one window is open.
frame	boolean	no	Include the native window frame, menus and dialogs

Clicks a point of the window content (Input.dispatchMouseEvent). count=2 is a double click.

Parameter	Type	Required	Notes
x		yes	—
y		yes	—
button	string	no	Mouse button
count	integer	no	Click count (2 = double click)
modifiers	array	no	Modifier keys held during the action (meta = ⌘ on macOS)
window	string	no	Window: target id from list_windows or part of its title. Optional when only one window is open.

Presses the left button at one point, moves to another with intermediate steps and releases (a real drag).

Parameter	Type	Required	Notes
x1		yes	—
y1		yes	—
x2		yes	—
y2		yes	—
duration_ms	integer	no	Drag duration in ms
modifiers	array	no	Modifier keys held during the action (meta = ⌘ on macOS)
window	string	no	Window: target id from list_windows or part of its title. Optional when only one window is open.

Moves the mouse to a point without clicking (hover effects, tooltips).

Parameter	Type	Required	Notes
x		yes	—
y		yes	—
window	string	no	Window: target id from list_windows or part of its title. Optional when only one window is open.

Scrolls with the mouse wheel at a point. Positive delta_y scrolls down; deltas are in CSS pixels.

Parameter	Type	Required	Notes
x		yes	—
y		yes	—
delta_x	number	no	Horizontal delta in CSS px
delta_y	number	no	Vertical delta in CSS px
window	string	no	Window: target id from list_windows or part of its title. Optional when only one window is open.

Types text into the focused element (Input.insertText; any Unicode). Click the field first; for Enter or shortcuts use press_key.

Parameter	Type	Required	Notes
text	string	yes	Text to type
window	string	no	Window: target id from list_windows or part of its title. Optional when only one window is open.

Presses a key, optionally with modifiers (e.g. key "a" + ["meta"] selects all on macOS). Named keys: …;

Parameter	Type	Required	Notes
key	string	yes	Key name or single character
modifiers	array	no	Modifier keys held during the action (meta = ⌘ on macOS)
window	string	no	Window: target id from list_windows or part of its title. Optional when only one window is open.

Sets the size of the window content area in CSS px. With the main process it resizes the real window;

Parameter	Type	Required	Notes
width	integer	yes	Content width in CSS px
height	integer	yes	Content height in CSS px
window	string	no	Window: target id from list_windows or part of its title. Optional when only one window is open.

Reloads the page of the window (renderer only; the main process keeps running).

Parameter	Type	Required	Notes
ignore_cache	boolean	no	Bypass the cache
window	string	no	Window: target id from list_windows or part of its title. Optional when only one window is open.

Evaluates a JavaScript expression in the window page (main world) and returns its JSON value. Promises are awaited.

Parameter	Type	Required	Notes
expression	string	yes	JavaScript expression
window	string	no	Window: target id from list_windows or part of its title. Optional when only one window is open.

Evaluates a JavaScript expression in the app's main process (Node inspector) and returns its JSON value.

Parameter	Type	Required	Notes
expression	string	yes	JavaScript expression
app	string	no	App: id (its CDP port) or name from list_windows. Optional with a single app.

Returns the accessibility tree of the window as indented text (role "name" value [state]). Interactive

Parameter	Type	Required	Notes
interactive_only	boolean	no	Only interactive elements, as a flat list
window	string	no	Window: target id from list_windows or part of its title. Optional when only one window is open.

Console messages and uncaught exceptions of the app: renderer windows and main process (up to the last 1000).

Parameter	Type	Required	Notes
level	string	no	Comma-separated levels to keep: log, info, warn, error, debug
source	string	no	Where the messages come from
limit	integer	no	Most recent entries to return
window	string	no	Only this window (plus the main process)
app	string	no	App: id (its CDP port) or name from list_windows. Optional with a single app.

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

Nothing here is billable: mcprush charges nothing to install Electron Viewer or to call it, at any volume.

ATrust score 93 / 100

Release 1.0.0 · scanned 3 Nov 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 score93

Findings1 threat · 0 capability

Capabilityhigh

Coveragesource

Verificationrepo

Surface digest0c635ab28c8f613887c0535126183a846bc9434051021fde6ea6bc9c1c1c5dc6

Score vectorthreat 100 → client 93

Term	Level	What it prices	Points
capability-exposure	high	capability blast radius (high) — client exposure if the model is manipulated	−6
verification-discount	repo	publisher verification (public source) — no provenance, but the source is public and inspectable	−1
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 (16 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
list_windows	no tags
launch_app	no tags
attach	no tags
screenshot	no tags
click	no tags
drag	no tags
hover	no tags
scroll	no tags
type_text	no tags
press_key	no tags
resize_window	no tags
reload	no tags
evaluate	no tags
evaluate_main	no tags
get_accessibility_tree	no tags
get_console_logs	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 digest0c635ab28c8f6138

Scanned2026-11-03

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.