EN中
Explore / Browserbase MCP
MCP ServerL3medium risk

Browserbase MCP

Run browser automation in hosted Browserbase sessions for agents that need web actions.

RepositoryDocs
Curated baseline · 52

A curated public baseline; every entry carries a verification level and risk labels. 25 from the automated pipeline

72
Recommended
Confidence: medium
79
Benchmark
Confidence: medium
Needs human reviewL3 · Seed profile normalized into AgentMaps schema.
Why
Recommendation 72, fits Browser, setup is medium.
Best for
Developers using Self-hosted for browser workflows.
Not for
Users expecting a fully managed marketplace install flow.
Boundary
Verified to L3; L4 not covered, so this is not full runtime proof.
Remaining risk
Risk appears manageable for personal developer workflows when configured narrowly.
Next step
Review tokens, write access, and network scope in a sandbox before team adoption.
Review before adoption
This capability includes high-risk, token, write, shell, or auth-gated gaps. L4 means interface parsing, not production safety approval.

What it is good for

Run browser automation in hosted Browserbase sessions for agents that need web actions. AgentMaps treats this as a mcp server candidate and scores it with a capped benchmark score plus a separate recommendation score that includes trust, platform fit, setup preference, and risk preference.

Use cases

  • - Navigate controlled websites
  • - Extract page data
  • - Operate browser workflows
  • - Run repeatable task flows
  • - Coordinate tool calls

Best for

  • - Developers using Self-hosted for browser workflows.
  • - Teams that want visible setup, verification, and risk evidence before adoption.

Not for

  • - Users expecting a fully managed marketplace install flow.
  • - Users who cannot provide the required credentials.

Limitations

  • - Scenario-level L5-L7 benchmark testing is not part of the current MVP record.
AI-native runtime contractTool/runtime connector

Runtime pattern before adoption

AI may help filter, explain, and draft a trial plan, but must not perform write, shell, external side effects, or production publishing before review.

Visible states

  • Collect task context and platform constraints
  • Preview setup, source, and permission boundaries
  • Trial in a sandbox or low-permission environment
  • Wait for human review before continuing

User controls

  • Revise task/platform filters
  • Add to Compare
  • Open source for review
  • Cancel high-risk adoption
  • Submit pending/staging evidence

Approval gates

  • Token/API key scope approval
  • External network target approval
  • Auth-gated path not fully tested

Failure recovery

  • Auth path needs manual reproduction or a constrained test account.

Trial acceptance

  • Can a user find a task-fit candidate within two minutes
  • Can the UI explain why it is recommended and where it does not fit
  • Can the user identify token, write, shell, network, or local file risk
  • Can the user separate L1/L2 evidence from full runtime proof
  • Does a failed trial have fallback or manual takeover

Verification evidence

The matrix shows passed, partial, skipped, and not-tested boundaries. It is not production adoption approval.

L1 Metadata

Source, docs, license, or package metadata exists.

passed
L2 Static audit

Static review assigns permission and risk boundaries.

passed
L3 Install path

Install path can be checked, but not necessarily in your environment.

passed
L4 Interface

Interface or entrypoint parsed; not a production safety approval.

skipped
  • Seed profile normalized into AgentMaps schema.
  • Source and documentation fields are present.
  • Static risk flags are assigned.
  • Install path is represented for harness checks.
  • Interface parsing pending.
  • Benchmark score capped at 80 by L3 verification.

Trust profile

Heuristic estimate
Needs reviewstaticTrigger risk medium

Risk findings

  • - Requires Token
  • - Hosted Browser
  • - External Network

Verified evidence

  • - Seed profile normalized into AgentMaps schema.
  • - Source and documentation fields are present.
  • - Static risk flags are assigned.
  • - Install path is represented for harness checks.

Score breakdown

Safety70
Execution proxy80
Maintenance82
Setup77
Interface85
Standards78
Task fit84
Portability85

Why it scores well

  • - Clear task fit for the selected scenario.
  • - Static verification evidence is available.
  • - Portable across multiple AI clients.

Watch outs

  • - Requires credentials before full testing.
  • - Production adoption still needs local validation.

Alternatives

GitHub MCP Server
MCP Server · Coding, Automation
66
Rec

Connect agents to GitHub repositories, issues, pull requests, and code search.

L4high riskSelf-hosted
View details →
Playwright MCP
MCP Server · Browser, Automation
84
Rec

Expose browser automation primitives through MCP for web navigation and testing.

L4medium riskSelf-hosted
View details →
Supabase MCP
MCP Server · Data, Coding +1
62
Rec

Give agents controlled access to Supabase projects, schemas, SQL, and project metadata.

L3high riskSelf-hosted
View details →