Does hyperbrowser-mcp send data, and where? — data-flow verdict

provisional · AUTOMATED — forensic confirmation pending. A preliminary, fact-based flag, not a judgment that the tool is unlawful or unsafe.

Verdict (the facts)

Tool

npm/hyperbrowser-mcp

Integrity axis

undisclosed_processing — Observed behaviour matches the tool's stated function; the egress above is the tool doing its advertised job. 'honest' is the integrity axis — it does NOT imply the data flow is irrelevant; see the data-flow axis and jurisdiction.

Data-flow axis

Sends data to app.hyperbrowser.ai (US, jurisdiction tier 2) as its core function. No telemetry, analytics or error-reporting side-channel was found (full source + dependency-tree review). Where your data goes is determined by the tool's stated purpose, not by a hidden observability channel.

Disclosure

functional — Single egress host, app.hyperbrowser.ai, and it is FUNCTIONAL — the cloud-browser API the tool exists to call. All 8 captured requests target functional paths (/api/extract, /api/task/browser-use, /api/task/cua, /api/task/claude-computer-use, /api/profile, /api/profiles), authenticated with x-api-key. No telemetry/analytics SDK.

Capture self-test

verified

Severity

low — integrity axis only (undeclared exfiltration). Functional egress and disclosed metadata are reported as neutral facts and are not graded here.

Version (pinned)

1.0.25 · commit

Content hash

sha256:ca6e556b244015131248c2692422ef6724ca89f37e32ab93f11807f993714619

Signature

ed25519:U1+/ON4GE9WtOLhhtdosRPdRok9MF2Dh/oHmp2… · Ed25519 public key · sha256:49cf8457b42a7048

Scanned

2026-06-14T00:00:00Z — Pinned to hyperbrowser-mcp@1.0.25, published 2025-04-13. This verdict applies to that exact version; a newer release would require a re-scan.

Re-verified

2026-06-14 — pinned version current

Categories

browser-automation functional-egress US published

Observation history

1 scan(s); first seen 2026-06-14T00:00:00Z · latest 2026-06-14T00:00:00Z

Observed egress destinations

Each destination is classified FUNCTIONAL (the tool's advertised job requires the call — a neutral fact about where your data goes), SESSION/AUTH (handshake with the same operator), or TELEMETRY/ERROR_REPORTING (an observability side-channel not required for the function). Disclosure is judged across the tool's full public doc surface, not just its README, and any 'undisclosed telemetry' finding is adversarially refuted before it is asserted.

Jurisdiction context: Tier 2 = third country (e.g. US): transferring EU personal data to a third country requires a transfer basis under GDPR Art. 44-49 (e.g. SCCs / EU-US Data Privacy Framework) — an obligation on you, the deployer; the tool gives no control over this flow. This is the applicable framework, not a finding that the tool violates it.

Evidence — the captured request (verify, don't just trust)

Capture self-test: verified — a beacon decoy was emitted from the tool's network context; its presence in the intercept means a 'no egress' result would have been trustworthy.

Observed: POST https://app.hyperbrowser.ai/api/extract ×8 — intercepted (the tool's HTTPS was terminated against the sandbox CA; the egress was then blocked by strict-egress, but the full request was captured)

Payload fields actually sent:

• urls
• sessionOptions.useProxy
• sessionOptions.useStealth
• sessionOptions.solveCaptchas
• sessionOptions.acceptCookies
• prompt
• task
• maxSteps
• sessionOptions.adblock
• schema.type
• schema.properties.allSearchResults.type
• schema.properties.allSearchResults.items.type
• schema.properties.allSearchResults.items.properties.title.type
• schema.properties.allSearchResults.items.properties.title.description
• schema.properties.allSearchResults.items.properties.url.type
• schema.properties.allSearchResults.items.properties.url.description
• schema.properties.allSearchResults.items.properties.snippet.type
• schema.properties.allSearchResults.items.properties.snippet.description
• schema.properties.allSearchResults.items.required
• schema.properties.allSearchResults.items.additionalProperties
• schema.required
• schema.additionalProperties

Captured payload sample (one event):

{"urls":["https://www.bing.com/search?q=FILE-CONTENT::canary-e525e270-file-123d591ce014::END"],"sessionOptions":{"useProxy":false,"useStealth":false,"solveCaptchas":false,"acceptCookies":false,"adblock":true},"prompt":"Extract the top 10 search results from this page.","schema":{"type":"object","properties":{"allSearchResults":{"type":"array","items":{"type":"object","properties":{"title":{"type":"string","descriptio

Captured in the sandbox run. The distinct_id (a persistent machine identifier) and the write-only, public-by-design ingestion key are truncated above; payload_fields is the union observed across the run.

Reproduce it yourself (canary-sandbox (open methodology; Docker backend)):
python -m canary.cli scan <target> --backend docker # target: npm hyperbrowser-mcp@1.0.25
Re-run it yourself: the scanner installs the pinned version, drives the tool over MCP, and intercepts all egress.

Full raw captured trace + verification: /verdict/hyperbrowser/evidence.json — every captured request (redacted), the verdict content-hash and the package checksum, for an AI or auditor that wants the underlying observation, not just the conclusion.

Disclosure check (the §824 evidence)

Read

package source (single egress host); docs (cloud service, API key required); captured sandbox requests

Quoted from the tool's own docs

“”

Match

Single egress host, app.hyperbrowser.ai, and it is FUNCTIONAL — the cloud-browser API the tool exists to call. All 8 captured requests target functional paths (/api/extract, /api/task/browser-use, /api/task/cua, /api/task/claude-computer-use, /api/profile, /api/profiles), authenticated with x-api-key. No telemetry/analytics SDK.

Residual gap

By design this is a CLOUD browser: the pages and content you browse leave to Hyperbrowser's cloud (US) — an inherent data-flow of using a cloud-browser service, not exfiltration. The docs state it is a cloud service requiring an API key (disclosure of the cloud nature: partial — the by-design content egress is implied, not spelled out).

How we know this — claims by basis

A verdict is a reproducible evidence container, not just a claim. Each assertion is tagged: an observation is in the capture and reproducible; an inference is our reasoning over it; documented is the tool’s own statement; a classification is our adversarially-reviewed judgment. Observation never reads as inference.

Observed — directly in the capture, reproducible

• The tool sent 8 request(s) to app.hyperbrowser.ai carrying fields: urls, sessionOptions.useProxy, sessionOptions.useStealth, sessionOptions.solveCaptchas, sessionOptions.acceptCookies, prompt, task, maxSteps, sessionOptions.adblock, schema.type, schema.properties.allSearchResults.type, schema.properties.allSearchResults.items.type. — Captured in the sandbox run (published redacted in the evidence artifact); re-run the scan to reproduce. (confidence: high)

Inferred — our reasoning over the observation

• The repeated requests suggest the flow fires on launch and on each tool call. — 8 requests in one run — an inferred pattern, not proven across launches. (confidence: medium)

Documented — the tool's own statement

• The tool's own docs state (quoted): ist/server.js ``` ## Claude Desktop app This is an example config for the Hyperbrowser MCP server for the Claude Desktop client. ```json { "mcpServers": { "hyperbrowser": { "command": — npm registry readme for hyperbrowser-mcp (confidence: high)

Classified — our adversarially-reviewed judgment

• app.hyperbrowser.ai is classified as egress (an observability side-channel, not required for the function). — Adversarially reviewed. (confidence: high)
• Disclosure status: disclosed. — The observed endpoint app.hyperbrowser.ai is named/disclosed in the tool's own docs (matched: app). (confidence: high)

Method

Installed and run in an isolated container; fed traceable decoy data; all outbound traffic intercepted (TLS broken via own CA, iptables transparent redirect). Endpoints, resolved geo/jurisdiction and frequency are observed facts. Capture self-test passed.

Scope

Compares the tool's declared destinations against what was observed in one sandbox run. Checks transparency / integrity for a cooperative tool, NOT resistance to deliberate evasion. "honest"/"clean" means "observed without deviation within our reach", NOT "guaranteed no hidden egress". Out of scope: exfiltration split/chunked across requests; tool-side encryption of the payload before egress; input/time/state-triggered processing not triggered in the run.

Machine-readable verdict: /verdict/hyperbrowser.json. This page describes observed behaviour and its relation to the tool's own disclosures — it is not a legal judgment. Search context: does hyperbrowser-mcp send data, hyperbrowser-mcp privacy, hyperbrowser-mcp data flow, hyperbrowser-mcp telemetry, where does hyperbrowser-mcp send data, is hyperbrowser-mcp safe, what data does hyperbrowser-mcp collect, how to disable hyperbrowser-mcp telemetry, hyperbrowser-mcp opt out tracking, hyperbrowser-mcp GDPR data residency, hyperbrowser-mcp third-party / jurisdiction.