Perplexity Browser MCP
NPM · PERPLEXITY-BROWSER-MCP
Search Perplexity using your browser login. Returns answers, sources, and thread URLs. No API key.
Available components
Last scanned (13 hours ago)
Scores higher than 36% of npm servers
Category Search and Retrieval
How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →
Supply Chain Security98
- No malware found by supply-chain analysis.Pass
- No known CVEs affecting this package version or its production dependencies.Pass
- No install/post-install scripts declared.Pass
- 31 of 95 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency19
- Repository check failed: no source repository is declared. See how to fix → View diagnostics → Fail
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- Clear OSI-approved license (MIT).Pass
- Actively maintained (last published 7 days ago).Pass
- Security-disclosure policy not yet verified: we couldn't inspect the source repository.Unverified
Schema Quality & AI Usability71
- AI-judged instruction clarity (good).Pass
- Tool/resource definitions use about 945 tokens (~67/item across 14 items; 14 tools + 0 resources), lean.Pass
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management0
- Stability not yet verified: not enough scan history yet (needs a 30-day window).Unverified
Tool Coverage67
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 0% of tool parameters carry a description.Fail
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- All 3 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.Pass
- An AI judge read all 15 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
Unverified: 1 category
A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.
Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 14 Oct 26 +28
- Injection markers: unverified → pass ▲ security
- First check of Judged manipulation: pass security
- Stability: Stability not yet verified: not enough scan history yet (needs a 30-day window). security
- Tool coverage: unverified → 100 ▲ functional
- MCP protocol: unverified → pass ▲ functional
- First check of Tool coverage: 0 functional
- First check of Schema quality: fail functional
- First check of Destructive annotations: 100 functional
- First check of Schema quality: pass functional
- First check of Schema quality: good functional
- 11 Oct 26 33
First indexed and scored.
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 18 Oct 2026 · Analysed npm/perplexity-browser-mcp@1.4.0
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | npm |
Background: How many MCP packages publish verified provenance →
Dependencies 95 packages
| Packages resolved | 95 |
|---|---|
| Stale | 31 |
| Tree resolution | Complete |
Background: SBOMs and build attestations, explained →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
ask_prompt ~34
Read or refresh the current Perplexity Ask prompt and handoffId. No model is called and no browser action is executed.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
ask_reply ~70
Execute one user-pasted Perplexity Ask JSON response for the current handoffId. Supports JSON fences and multiline JSON. Returns the updated snapshot and next prompt. Never fabricate the response; wait for the user to paste it.
| Name | Type | Req | Description |
|---|---|---|---|
| handoffId | string | yes | – |
| response | string | yes | – |
No output schema declared.
No examples provided.
ask_start ~64
Start a browser task planned through Perplexity Ask with manual copy/paste. Returns a prompt to show the user; wait for their actual Ask reply before executing any action.
| Name | Type | Req | Description |
|---|---|---|---|
| demo | boolean | – | – |
| goal | string | yes | – |
| url | string | – | – |
No output schema declared.
No examples provided.
browser_action ~54
Execute one local browser action and return the updated snapshot. Match accessible roles/names from the snapshot. Actions count toward the session step limit. done records the planner outcome; it does not independently verify success.
| Name | Type | Req | Description |
|---|---|---|---|
| command | – | yes | – |
No output schema declared.
No examples provided.
browser_close ~46
Close the owned task browser and reset its step limit. Chromium discards task cookies; Arc and Safari keep your normal account login. The separate Perplexity Ask window remains until web_close.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
browser_screenshot ~44
Return a PNG image of the current viewport in Chromium mode. Arc and Safari native scripting do not support screenshots. Use only when text snapshots are insufficient; no file is written.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
browser_snapshot ~49
Read the current page snapshot (up to 8000 characters), tabs, steps, and recent results. Arc and Safari use DOM role/name mapping; Chromium uses its accessibility tree. Call browser_start first.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
browser_start ~64
Start a local task browser using the selected Chromium, Arc, or Safari adapter. Choose url or demo=true. Native browsers control a dedicated tab and uses a DOM role/name snapshot. Close before starting another session.
| Name | Type | Req | Description |
|---|---|---|---|
| demo | boolean | – | – |
| url | string | – | – |
No output schema declared.
No examples provided.
search ~84
Ask Perplexity a question directly on its website. Automatically connect, submit the exact query, and wait for the complete answer. Return DOM answer text, deduplicated source links, and the thread URL. Use this for research, not web_start/web_step. May take up to 90 seconds; sign in in the opened window if required.
| Name | Type | Req | Description |
|---|---|---|---|
| query | string | yes | – |
No output schema declared.
No examples provided.
web_close ~36
Close both owned browser windows and reset workflows. Website login persists in the dedicated profile or existing native browser account. Does not close unrelated windows.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
web_connect ~50
Open Perplexity Ask in a dedicated window. The user signs in directly in the browser. Arc and Safari reuse the browser account; Chromium uses a separate persistent profile. Returns connection status without reading account history.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
web_start ~68
Open a target page with a goal for automatic Perplexity Ask website planning. Call web_connect and complete sign-in first. Call web_step to send a prompt and execute one actual website reply.
| Name | Type | Req | Description |
|---|---|---|---|
| demo | boolean | – | – |
| goal | string | yes | – |
| url | string | – | – |
No output schema declared.
No examples provided.
web_status ~33
Check whether the dedicated Perplexity Ask window is signed in and ready. Does not submit a prompt or return account history.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
web_step ~69
Send the current target snapshot and goal to the real Perplexity Ask website, wait for its complete validated JSON reply, execute one action, and return the updated target snapshot. Stops without executing if the reply is invalid, times out, or the target snapshot changed. May take up to 90 seconds.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
How do I install the Perplexity Browser MCP server?
Perplexity Browser MCP runs locally as an npm package, launched with npx -y perplexity-browser-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
npm · perplexity-browser-mcp
claude mcp add xs1128-perplexity-browser-mcp -- npx -y perplexity-browser-mcp
What is the Perplexity Browser MCP server?
Perplexity Browser MCP is listed in the public MCP registry as io.github.xs1128/perplexity-browser-mcp. Search Perplexity using your browser login. Returns answers, sources, and thread URLs. No API key. This page covers its npm package (perplexity-browser-mcp).
Is the Perplexity Browser MCP server safe to use?
Perplexity Browser MCP scores 61 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 18 October 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.
What tools does the Perplexity Browser MCP server expose?
Perplexity Browser MCP exposes 14 tools: search, ask_start, ask_prompt, ask_reply, web_connect, and 9 more. Their descriptions and schemas cost roughly 765 tokens of context every time the server is loaded.
Is the Perplexity Browser MCP server still maintained?
Perplexity Browser MCP is still listed as active in the MCP registry. We last reached this channel on 18 October 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.
What licence is the Perplexity Browser MCP server under?
Perplexity Browser MCP declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.