Hyperbrowser

# Fix list: Hyperbrowser

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/hyperbrowser, the October 2026 research run, assessed 8 October 2026. Grade B, 66.9 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Hyperbrowser: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Security & auth, 44 out of 100, up to 9.8 more on the total

Why it scored 44: One `x-api-key` per account with no scopes or rotation guidance found (10). The x402 route needs no key and returns session-scoped tokens plus a `managementToken` that can only read or stop that session, which we credit 5. Connection tokens travel in the `wsEndpoint` URL and the MCP tools accept an `apiKey` argument, which we note without deducting because neither puts the account key in a documented URL. `viewOnlyLiveView`, a URL blocklist and sandbox network policies exist, with no read-only key (7). Pages are untrusted content and no prompt-injection guidance was found (0). rrweb recordings on by default, optional video and a live view give a per-session trail (11). SOC 2 Type 1 on request, annual third-party penetration tests and a Secureframe-monitored trust centre. No security.txt and no bug bounty found (11).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 2. Reliability, 65 out of 100, up to 7 more on the total

Why it scored 65: Graded as a hosted API. hyperbrowser.betteruptime.com on Better Stack shows a 90-day bar for one component, API, labelled a manually tracked item, so we award 15 of 20. It reads 100 per cent with no incidents listed in the 90 days to 8 October 2026, and we award 25 of 30 because the item is tracked by hand. Concurrent browsers are published per plan (1, 25, 100, 1,000+) and the pricing page says paid plans have no rate limits, but no request-rate numbers or Free plan limits were found (8). 429 is documented on batch scrape and x402 session capacity, the x402 guide explains which failed requests can be retried, and the Node SDK retries control-plane GETs on 429 and 5xx three times with backoff. No Retry-After header or idempotency keys found (7). No SLA found (0). The API is version 1.0.0 and both SDKs are 1.x (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 3. Agent ergonomics, 67 out of 100, up to 5.4 more on the total

Why it scored 67: Fetch takes output formats, include and exclude selectors and a cache age, Search takes a page and filters, and the MCP server has 10 tools (20). Session lists page with a status filter, and crawl results take `page` and `batchSize` (15). Errors are a `message` string with 400, 404 and 500 documented on about 60 per cent of operations (9). No idempotency keys on session creation. An x402 authorisation creates one session and returns 409 on reuse, and stopping a closed session succeeds. The MCP tools carry no annotations (8). Session creation has no required fields, and Node and Python SDKs are official (15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 4. Schema & documentation, 68 out of 100, up to 5.2 more on the total

Why it scored 68: Graded on the REST API. OpenAPI 3.0.1 at /docs/openapi.json with 87 paths and 91 operations (25). llms.txt on the site and in the docs, with a Markdown twin of every page (10). Only 12 of 91 operations and 106 of 707 schema properties carry a description, though the guides explain each product well (9). 170 schemas with 26 enums and ranges such as `timeoutMinutes` 1 to 720, and session creation has no free-form blobs apart from `browserArgs` (10). JavaScript and Python samples on most operations. 35 of 91 operations document no error response and the error body is a single `message` string (8). The spec is 1.0.0 with no version in the path, and the website changelog's newest entry is 18 November 2025, so dated changes are only visible through SDK tags (6).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 5. Transparency & trust, 64 out of 100, up to 3.2 more on the total

Made of editorial 55, provenance 72.

Why it scored 64: The platform is closed with published terms, the SDKs and MCP server are MIT and HyperAgent is AGPL-3.0 (20). The privacy notice of 24 October 2025 covers only the website. Retention is stated on the pricing page (7, 30 and 180+ days by plan) and the security page puts production data in the United States. No public DPA found (14). No deprecation policy found. The Python 1.0 migration guide keeps earlier request objects valid (5). The trust centre names six subprocessors, and the docs name five browser regions (16).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- Domain age: hyperbrowser.ai, registered 2024-11-19 (1 year) (3 of 15)
- Terms of service: published, but our reader couldn't read it (7 of 10)
- Privacy policy: published, but our reader couldn't read it (7 of 10)
- security.txt: not found (0 of 10)

## 6. Maintenance & community, 79 out of 100, up to 1.8 more on the total

Why it scored 79: Node SDK 1.1.0 tagged 7 October 2026 and Python SDK 1.10.0 uploaded to PyPI on 8 October (30). 25 PyPI releases since 10 July 2026 (20). The website changelog stops at 18 November 2025, support is a Discord server, and we couldn't read GitHub issue reply times (7 of 15 on the closed-service line). Current official SDKs in Node and Python. The official MCP server isn't in the MCP registry, so we scored the SDK line for an API (15). Both SDK repositories run integration tests in CI, but hyperbrowser-mcp was last published on 13 April 2025 and pins SDK ^0.48.1 (7).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## 7. Payments & pricing, 94 out of 100, up to 0.8 more on the total

Why it scored 94: x402 version 2 on the vendor's own host for sessions, Fetch and Search, with USDC on Base and Solana. Unpaid requests on 8 October 2026 returned 402 with terms. Scrape, Crawl, Extract, agent tasks and sandboxes need a key, so partial (34). Per-unit prices published without login (20). Free plan with 5,000 credits, and the pricing page says no card is required (20). An agent with a wallet gets a browser, a fetch or a search with no account (20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- unchecked: GitHub stars and open-issue reply times. The GitHub API refused us for its rate limit, so `githubStars` is empty and responsiveness is scored from the changelog and support channels only
- unchecked: the SOC 2 Type 1 report itself and any DPA, which the trust centre releases on request
- unchecked: whether the dashboard allows several API keys, revocation or scopes. The public docs don't say and the dashboard needs a login
- unchecked: the x402 price of a Search query. We probed sessions and Fetch only, and the docs say the price is calculated from the request options
- The website pages (pricing, terms, privacy, security, changelog) render in the browser, so we read their text from the page's own script files rather than rendered HTML
- Whether the Free plan's 5,000 credits are one-off or monthly. The pricing FAQ says they are for testing the service and that Free users must upgrade when they run out
- No date was found on the terms of service
- The status page's one component is labelled a manually tracked item, so the clean 90-day record may not reflect automated monitoring

## Weaknesses

- The official MCP server's last npm release is 1.0.25 of 13 April 2025, and it isn't in the official MCP registry
- No request rate limits, Retry-After header or SLA found; the pricing page says paid plans have no rate limits
- One `x-api-key` per account with no scopes or rotation guidance found in the reviewed documentation
- The privacy notice of 24 October 2025 covers only the website, and no DPA is public
- The website changelog stops at 18 November 2025, and 35 of 91 OpenAPI operations document no error response

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- With a wallet and no key, POST to https://api.hyperbrowser.ai/x402/session and read the `PAYMENT-REQUIRED` header. Sessions are prepaid for 5 to 60 minutes with no refund
- Save the x402 creation response. It holds the only copy of the connection tokens, and `managementToken` is what stops the session
- Stop sessions explicitly in cleanup code. Browser time bills per second until the timeout
- Use Fetch at $0.001 a page for static pages and keep browser sessions for pages that need interaction
- Add `&keepAlive=true` to `wsEndpoint` to reconnect after a CDP disconnect. Treat that URL as a secret because it carries the session token

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.