Local-first knowledge graph for token-efficient code review through MCP and CLI
Nothing to run from here
Nerve 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.
89/100
trust score
42
downloads in 30d
local
runs on your machine
40
tools declared
What it does
MCP server for Nerve — gives AI agents runtime access to iOS apps
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user nerve-mcp '--' npx -y nerve-mcp
# 2 — ask your agent for something these tools do
# nerve_view, nerve_tree, nerve_inspect and 37 more
Nerve costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
LUAll 1 server
luchi02081 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 Nerve 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 nerve-mcp '--' npx -y nerve-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
40 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.
See the current screen. Returns all visible UI elements with their type (btn, txt, field, toggle), label, identifier (#id), and position. This is your primary tool for understanding what
No parameter schema on file.
Dump the complete view hierarchy tree showing all views with nesting, types, and frames.
No parameter schema on file.
Inspect a specific UI element. Returns properties, accessibility info, constraints, and type information. Query by #identifier, @label, .Type, or x,y coordinates.
No parameter schema on file.
Tap a UI element. Response includes the updated screen state (auto-view), so you do NOT need to call nerve_view after tapping. Use #identifier, @label, or x,y coordinates.
No parameter schema on file.
Scroll the current scroll view in a direction.
No parameter schema on file.
Perform a swipe gesture in a direction.
No parameter schema on file.
Double-tap a UI element. Used for zoom in/out on maps, text selection.
No parameter schema on file.
Drag and drop: long-press the source element, drag to the target, release.
No parameter schema on file.
Pull down to refresh the current scroll view content.
No parameter schema on file.
Two-finger pinch/zoom gesture. Scale > 1.0 zooms in, < 1.0 zooms out.
No parameter schema on file.
Open a context menu by long-pressing an element. Returns the current screen showing menu items. Use nerve_tap to select a menu item.
No parameter schema on file.
Long-press a UI element. Triggers context menus, drag initiation, and haptic touch actions.
No parameter schema on file.
Type text into the currently focused text field. Tap the field first with nerve_tap to focus it, then call this to enter text.
No parameter schema on file.
Navigate back (pop navigation or dismiss presented view controller).
No parameter schema on file.
Dismiss the keyboard or the frontmost presented view controller.
No parameter schema on file.
Capture a screenshot. WARNING: Screenshots consume significant tokens — use nerve_view for all normal inspection. Only use for visual checks (colors, gradients, images, layout). Use
No parameter schema on file.
Read the app
No parameter schema on file.
Show recent HTTP requests with method, URL, status, and timing. In-flight requests show as
No parameter schema on file.
Find live instances of a class on the heap and inspect their properties. First call with just class_name to list instances. Then call with index to read all properties of a specific instance (e.g., check a ViewModel
No parameter schema on file.
Read app storage: UserDefaults, Keychain, cookies, sandbox files, or Core Data. For Core Data, omit entity to list all entities, or specify entity to fetch records.
No parameter schema on file.
Check connection status, app state, and Nerve version for all connected targets.
No parameter schema on file.
Show the app
No parameter schema on file.
Auto-navigate to a screen by name (e.g.,
No parameter schema on file.
Invoke a custom accessibility action on an element. Use nerve_inspect to discover available actions first.
No parameter schema on file.
Find an element that
No parameter schema on file.
Wait until ALL network requests complete and animations finish. Simple but may wait too long if the app has background polling. For precise control, use nerve_network instead to check if a specific request completed.
No parameter schema on file.
Build an iOS app for the simulator using xcodebuild.
No parameter schema on file.
Build, install, and launch an iOS app on the simulator. Nerve is auto-injected if the app doesn
No parameter schema on file.
Open a deeplink URL in the app to navigate directly to a screen. Works with custom URL schemes (e.g.,
No parameter schema on file.
Pre-grant iOS permissions so system dialogs never appear during automation. Call before interacting with features that require permissions (camera, location, photos, etc.).
No parameter schema on file.
List available iOS simulators and their state (Booted/Shutdown).
No parameter schema on file.
Boot a simulator by name or UDID. Opens the Simulator app if not already open.
No parameter schema on file.
Switch between light and dark mode. On simulator, uses simctl. On device or if Nerve is connected, overrides the app
No parameter schema on file.
Build, install, and launch an iOS app on a connected physical device. Requires the device to be paired and a valid code signing identity.
No parameter schema on file.
List connected physical iOS devices and their state (available/unavailable).
No parameter schema on file.
Trace method calls at runtime via swizzling. Logs every invocation to the console (read with nerve_console). Zero overhead compared to LLDB breakpoints. Use this to understand code flow without rebuilding.
No parameter schema on file.
Draw a colored border around a UI element for visual debugging. Use with nerve_screenshot to see the result. Call with action=
No parameter schema on file.
Modify a UI element
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
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 Nerve 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: —
BTrust score 89 / 100
Release 0.2.2 · scanned 14 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 score89
Findings1 threat · 7 capability
Capabilitycritical
Coveragesource
Verificationrepo
Surface digest8c47fba63673cfde2e9ee23630f53d53e27512ac7b0275a3658527b8e118d5c2
Score vectorthreat 100 → client 89
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | critical | capability blast radius (critical) — client exposure if the model is manipulated | −10 |
| 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 (40 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. critical is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| nerve_view | no tags | |
| nerve_tree | sensitive-source | sensitive-source: keyword "dump" |
| nerve_inspect | no tags | |
| nerve_tap | no tags | |
| nerve_scroll | no tags | |
| nerve_swipe | no tags | |
| nerve_double_tap | no tags | |
| nerve_drag_drop | no tags | |
| nerve_pull_to_refresh | no tags | |
| nerve_pinch | no tags | |
| nerve_context_menu | no tags | |
| nerve_long_press | no tags | |
| nerve_type | no tags | |
| nerve_back | no tags | |
| nerve_dismiss | no tags | |
| nerve_screenshot | no tags | |
| nerve_console | no tags | |
| nerve_network | external-sink | external-sink: keyword "http_request" |
| nerve_heap | no tags | |
| nerve_storage | untrusted-input | untrusted-input: keyword "fetch" |
| nerve_status | no tags | |
| nerve_map | no tags | |
| nerve_navigate | no tags | |
| nerve_action | no tags | |
| nerve_scroll_to_find | no tags | |
| nerve_wait_idle | no tags | |
| nerve_build | no tags | |
| nerve_run | no tags | |
| nerve_deeplink | no tags | |
| nerve_grant_permissions | no tags | |
| nerve_list_simulators | no tags | |
| nerve_boot_simulator | no tags | |
| nerve_appearance | no tags | |
| nerve_run_device | no tags | |
| nerve_list_devices | no tags | |
| nerve_trace | no tags | |
| nerve_highlight | no tags | |
| nerve_modify | no tags | |
| nerve_lldb | code-exec | code-exec: parameter "command" |
| nerve_sequence | code-exec | code-exec: parameter "command" |
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.
nerve_storage → nerve_tree → nerve_lldb
A cross-tool exfiltration chain exists: nerve_storage → nerve_tree → nerve_lldb.
Supply chain and provenance
Integrityfirst-seen
Surface digest8c47fba63673cfde
Scanned2026-10-14
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 0.2.2. Pin a release below and the command asks for exactly that one.
0.2.2
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 Nerve 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.