Local MCP capabilities provider for Expo
Nothing to run from here
IsolatedTester — TinkyVision system 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.
93/100
trust score
17
tools declared
local
runs on your machine
121
downloads
What it does
Isolated macOS app tests, MCP images, OCR/ASCII frames, accessibility and evidence reports.
Quickstart
This one is not published to a package registry, so there is no one-line install: it is built and started the way its own source says.
Publishernot claimed
AGAll 1 server
AgewellEPM1 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 IsolatedTester — TinkyVision system 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.
This one is not published to a package registry, so there is no one-line install: it is built and started the way its own source says.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Capture the screen or one app window and return an MCP image directly to the connected AI client. The image may be sent to the client model provider. Requires Screen Recording permission.
No parameter schema on file.
Read the GhostBridge handoff phase plus active right-portal session identity, source kind, lifecycle/health, frame freshness, focus owner, cursor/keyboard capture, pressed input state, and TinkyStream paths. Read-only. Use this before reasoning about or controlling the portal.
No parameter schema on file.
See the active GhostBridge workspace through its verified composed left+right frame or TinkyStream full physical frame. The composed image is SHA-256 matched to atomic metadata and returned with session health/freshness. Read-only.
No parameter schema on file.
Read the live Perslis-to-remote-Mac control binding: exact source, paired agent, authenticated transport session, grant owner/expiry, and whether this MCP process owns the active grant. Read-only.
No parameter schema on file.
Ask the user for a visible, time-bounded grant allowing this MCP process to control the explicitly connected remote Mac in the GhostBridge right panel. The grant is bound to the paired agent and current authenticated transport session; reconnect revokes it.
No parameter schema on file.
Control the explicitly connected remote Mac after portal_remote_begin. Coordinates are in the direct portal source frame returned by portal_snapshot, not full-screen coordinates. Supported actions: move, click, scroll, key, type.
No parameter schema on file.
Move regular files through the authenticated remote-Mac session. send accepts absolute local paths and commits them into the remote owner-only transfer folder; receive accepts one safe filename from that folder and returns its verified local path.
No parameter schema on file.
Immediately release all remote mouse buttons and keys, return focus ownership to the left workspace, and revoke this MCP process control grant. This safety action remains available in read-only mode.
No parameter schema on file.
Capture a screenshot of the macOS screen or a specific app window. Returns the file path of the PNG. Use this to SEE what is currently on screen before reasoning about clicks. Safe — no consent prompt; screenshots are read-only.
No parameter schema on file.
List all running .app processes with their bundle IDs and PIDs. Use this to find the bundle ID of the app you want to control.
No parameter schema on file.
Search visible windows by title or owner-app substring. Returns matches with windowID, title, owner app, PID, and bounds (x/y/w/h). Use to locate a specific document window across multiple open instances.
No parameter schema on file.
Synthetic mouse click at SCREEN coordinates (x, y in pixels). REQUIRES Accessibility permission for the tinky-os helper binary. WRITE action — first call per target prompts the user for consent via macOS dialog. Use os_screenshot first to find coordinates.
No parameter schema on file.
Type a string of text at the currently-focused field. REQUIRES Accessibility permission. WRITE action — first call per target prompts the user for consent. Click into the target field first via os_click.
No parameter schema on file.
Press a single key with optional modifiers. REQUIRES Accessibility permission. WRITE action — first call per target prompts for consent. Use for keyboard shortcuts (Cmd+S, Cmd+W, Return, Escape, arrow keys, F1-F12).
No parameter schema on file.
Check whether the tinky-os helper has Accessibility permission. Returns { accessibility: true|false }. Call this first if click/type/key are failing — without Accessibility, they silently no-op at the OS level.
No parameter schema on file.
Run on-device OCR over the current screen (or a provided PNG) and return text matches with bounding boxes. Returns
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 IsolatedTester — TinkyVision system 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 93 / 100
Release 1.2.1 · 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 score93
Findings5 threat · 3 capability
Capabilityhigh
Coveragesource
Verificationrepo
Surface digest22b666e5409a880282d57c0c250bc34445ef271f5b1ca5e2e89c4b112622437f
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 (17 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 |
|---|---|---|
| os_screenshot_image | no tags | |
| portal_state | no tags | |
| portal_snapshot | no tags | |
| portal_remote_status | no tags | |
| portal_remote_begin | no tags | |
| portal_remote_control | no tags | |
| portal_remote_files | no tags | |
| portal_remote_release | no tags | |
| os_screenshot | no tags | |
| os_list_apps | no tags | |
| os_find_window | no tags | |
| os_click | no tags | |
| os_type | no tags | |
| os_key | code-exec | code-exec: parameter "cmd" |
| os_ax_check | no tags | |
| os_focused_window | no tags | |
| vision_find_text | 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 digest22b666e5409a8802
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
It installs from its repository rather than from a versioned package, so the command cannot be pinned to a release. The history below is for reading; the current release on file is 1.2.1.
1.2.1
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 IsolatedTester — TinkyVision system 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.