browser-relay
MCP server that drives a user's real Chrome browser via a WebSocket-connected MV3 extension, enabling tab management, navigation, page interaction, screenshots, and script evaluation through natural language.
Caution. Specific findings reduced this grade. They are listed on the page. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.
How we verified
Verified is a snapshot. Live keeps it current, and builds your track record.
⚡ Connect GitHub → continuous verification on every pushwhy connect →Who stands behind it
Wecko-ai
Source: Glama
Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find. Or connect your repo for our deepest verification, Live Monitored: read-only, revoke anytime. What we access →
Install from
The grade above is for the source repository. Registries can serve a different version, so we mark the ones we were not able to read.
These names and descriptions are the publisher's own, read from the source code. We print them as written. Our assessment is the findings above, not this list.
list_tabsList every open tab in the real Chrome, with id, title, url and which one is active. Start here to find the tab you need.
new_tabOpen a new tab, optionally at a url, and wait for it to finish loading.
close_tabClose a tab by id.
activate_tabBring a tab to the foreground and focus its window.
navigateNavigate a tab to a url and wait for load completion. Returns the final url after redirects.
backGo back in a tab history.
forwardGo forward in a tab history.
snapshotRead a page: title, url, frames, and interactive elements each with a "ref" id (iframes included, refs are prefixed per frame). Use those refs with click and type_text. Refs are invalidated by navigation, so re-snapshot after the page changes. Output is capped.
clickClick an element by the ref returned by snapshot.
type_textType text into an input, textarea or contenteditable identified by a snapshot ref.
evaluateRun JavaScript in the page (MAIN world) and return its JSON-serializable result, capped at ~6KB (strings truncated, arrays/objects limited). Pages with a strict Content-Security-Policy reject eval; prefer snapshot when possible. Return minimal data: booleans, short strings, small arrays.
screenshotCapture the visible area of a tab as an image, downscaled to max 1280px (jpeg 75 by default; png is larger and more expensive). The tab is activated for the capture and the previous tab restored. Vision-capable models pay tokens per image: use sparingly, prefer snapshot.
wait_forPoll a tab until the given text appears in the body of any frame (main page or iframe). Polls every 1.5s.
Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.
ws: Memory exhaustion DoS from tiny fragments and data chunks
ws: Uninitialized memory disclosure
BROWSERX_PORTTool annotations
No tools have read-only/destructive annotations
Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.
All four hints declared on every tool
13/13 tools missing one or more hints — list_tabs (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); new_tab (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); close_tab (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +10 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.
For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.
Production dependencies are patched
0 critical, 1 high severity in production deps — ws@8.18.0 (high), ws@8.18.0 (low)
Run npm audit fix, or upgrade the affected packages to a non-vulnerable version.
Claim the listing to review these findings one by one and send us a correction where you disagree, straight to the team. Claiming also means we tell you when the grade moves, and reach you first if we find anything urgent.
[](https://m8ven.ai/mcp/wecko-ai/browser-relay)?variant=verified from the URL.Vetting this one by hand? Tool Check is an MCP that scores other MCPs. Add it once and ask Claude, ChatGPT, or any MCP client to grade a server, surface CVEs, check the publisher, and suggest safer alternatives — before you install.
https://m8ven.ai/api/mcp/tool-check