MCP Server for Z.AI - A Model Context Protocol server that provides AI capabilities
Nothing to run from here
Sidebutton 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.
83/100
trust score
34
tools declared
local
runs on your machine
—
downloads a day
What it does
Open-source MCP server with browser automation and knowledge packs for AI agents
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user sidebutton-mcp '--' npx -y @sidebutton/server
# 2 — ask your agent for something these tools do
# run_workflow, get_run_log, list_workflows and 31 more
Sidebutton costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
SIAll 2 servers
sidebutton2 servers
2 categories
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 Sidebutton 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 sidebutton-mcp '--' npx -y @sidebutton/serverReconnect, 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
34 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.
Execute a workflow automation by ID. Returns a run_id for tracking execution status.
No parameter schema on file.
Retrieve the execution log for a completed workflow run.
No parameter schema on file.
List all available workflow automations.
No parameter schema on file.
Get detailed information about a specific workflow.
No parameter schema on file.
List recent workflow execution logs.
No parameter schema on file.
Publish a piece of evidence (a screenshot, a design mockup, an RCA/coverage report) to the portal RIGHT NOW, mid-session, and get back a paste-ready snippet to cite in your ONE resolution comment on the ticket. The file is uploaded, attached to the ticket automatically, and given a tokenized…
No parameter schema on file.
Check if the browser extension is connected.
No parameter schema on file.
Capture selectors and interactive elements from the current page.
No parameter schema on file.
Navigate the connected browser tab to a URL.
No parameter schema on file.
Capture accessibility snapshot of the current page. Returns YAML with element refs for use with click/type. Use includeContent=true to also include visible text content as markdown. Note: taking a snapshot may dismiss inline modals or popups — use screenshot instead if you need to verify modal…
No parameter schema on file.
Click an element on the page.
No parameter schema on file.
Press a key on the keyboard. Use this for keyboard shortcuts, Tab navigation, Enter, Escape, arrow keys, etc. Supports key combinations like "Ctrl+A" or "Shift+Tab".
No parameter schema on file.
Scroll the page.
No parameter schema on file.
Extract text content from an element.
No parameter schema on file.
Capture a screenshot of the current page. Prefer cropping to a specific area instead of capturing the full viewport — use ref (from snapshot), selector (CSS), or region (manual rect) to save context tokens. Full viewport is fine for first visit to a new page; after that, crop to the relevant…
No parameter schema on file.
Select an option from a native <select> dropdown element. Use this instead of click for <select> elements, as native dropdowns cannot be controlled via click events.
No parameter schema on file.
Fill a form field by setting its value programmatically. Unlike "type" (which simulates keystrokes), "fill" sets the value directly and triggers React/Vue/Angular change events. Use this for framework-controlled inputs where "type" doesn't work.
No parameter schema on file.
Wait for an element to appear on the page. Blocks until the element matching the selector exists in the DOM, or throws after timeout.
No parameter schema on file.
Check if an element exists on the page. Returns true/false without throwing. Useful for conditional logic.
No parameter schema on file.
Extract text from all elements matching a selector, joined by a separator. Useful for getting lists of items, table columns, or repeated elements.
No parameter schema on file.
Extract structured data from repeated elements. For each element matching the outer selector, extracts named fields using sub-selectors. Returns JSON array of objects.
No parameter schema on file.
Scroll a specific element into the viewport. More precise than "scroll" — targets an exact element.
No parameter schema on file.
Hover over an element (positions cursor for scroll targeting).
No parameter schema on file.
Inject a CSS rule into the current page. The main use is pre-capture redaction: blur or hide sensitive elements, then screenshot to a file, so no unredacted image ever exists. Pass an id to make the injection replaceable — injecting the same id again swaps the rule rather than stacking a second one.
No parameter schema on file.
Execute JavaScript in the browser page context and return the result. Useful for reading page state, checking values, or performing calculations.
No parameter schema on file.
Run a SEQUENCE of browser actions in ONE call instead of one round trip per action. Each step is the same shape you would pass to the individual tool, plus a "cmd" field naming which tool to run. A batch is a DETERMINISTIC run: actions you already know the selectors for, executed top to bottom,…
No parameter schema on file.
Provide HTTP Basic Auth credentials so the browser logs in silently instead of showing the native username/password modal, which would otherwise block the page and hang automation. Applies to the connected tab. Omit "origin" to use the credentials for any site that challenges; pass "origin" (e.g.…
No parameter schema on file.
Remove stored HTTP Basic Auth credentials. Omit "origin" to clear all stored credentials; pass "origin" to clear just that site.
No parameter schema on file.
Return the current counter value (proves state persists across calls).
No parameter schema on file.
Echo back the
No parameter schema on file.
Sleep for
No parameter schema on file.
Exit the process immediately — used to exercise restart-on-crash.
No parameter schema on file.
Add
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 Apache-2.0
- 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 Sidebutton 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 83 / 100
Release 1.5.9 · scanned 15 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 score94
Client score83
Findings5 threat · 2 capability
Capabilitycritical
Coveragesource
Verificationrepo
Surface digest4f78d43147f23ebfa3f2aa9559f24e3bd49a476825a47fb65a529fb81ab74c5c
Score vectorthreat 94 → client 83
| 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 (29 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 |
|---|---|---|
| run_workflow | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| get_run_log | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| list_workflows | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| get_workflow | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| list_run_logs | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| publish_artifact | untrusted-inputexternal-sink | untrusted-input: keyword "download"; external-sink: declares destructiveHint: true, openWorldHint: true |
| get_browser_status | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| capture_page | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| navigate | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| snapshot | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| click | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| type | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| press_key | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| scroll | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| extract | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| screenshot | no tags | |
| select_option | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| fill | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| wait | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| exists | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| extract_all | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| extract_map | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| scroll_into_view | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| hover | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| inject_css | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| evaluate | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| browser_batch | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| set_basic_auth | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| clear_basic_auth | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
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.
flow-1
Untrusted input can drive an external action even though no sensitive source is exposed.
Supply chain and provenance
Integrityfirst-seen
Surface digest4f78d43147f23ebf
Scanned2026-10-15
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 1.5.9. Pin a release below and the command asks for exactly that one.
1.5.9
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 Sidebutton 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.