Is Browserbase MCP server safe?
What Browserbase MCP server does
This server let an MCP client control cloud browsers on Browserbase, using Stagehand for model-driven page interaction. Its README now states that the repository is archived and no longer maintained, kept as a historical reference implementation that does not represent Browserbase's current production services. It required BROWSERBASE_API_KEY and BROWSERBASE_PROJECT_ID plus a model API key (Gemini by default), and offered an HTTP transport bound to a configurable host. Before installing, account for the fact that no further fixes will ship; if you still run it, keep --host at the default localhost rather than 0.0.0.0.
- MCP server launches a host command
- Node.js filesystem read
- Node.js filesystem write/delete
What to do: Nothing here argues against installing it. Grant the capabilities it lists only if you expect the tool to need them.
Browserbase MCP server is an MCP server analyzed by SkillTotal's deterministic static scanner. The scan found no malicious indicators. It can: filesystem read, filesystem write, install time execution, mcp tools detected and network egress — capabilities are what the code can do, not a verdict on intent. Risk score 0/100 (low).
@browserbasehq/ mcp 3.0.0
Get notified when this component changes
Change alerts aren't live yet. Leave your email and we'll let you know when they launch. We use it only for this and delete it on request.
Automated static-analysis result. It can contain false positives and false negatives, and is not a claim about the intent of Browserbase MCP server's authors. Report a false positive.
Behavioral traits
How this component maps to the CSA agentic threat model. Descriptive — it never affects the risk score.
Findings (6)
An MCP server entry launches a command on your host.
"command": "npx",
Why it matters: Trusting the manifest means running that binary — verify what it is and where it comes from.
Fix: Verify the launched command and its source before trusting this MCP server configuration.
The component reads files from disk.
const configContent = await fs.readFile(configPath, "utf-8");
const packageJSONBuffer = fs.readFileSync(packageJSONPath);
Why it matters: Usually legitimate, but worth confirming it can't be steered into reading sensitive files.
Fix: Confirm which files are read and that paths cannot be influenced by untrusted input to reach sensitive locations.
The component writes or deletes files on disk.
await fs.writeFile(wfPath, JSON.stringify(wfConfig, null, 2));
await Promise.all(workflowFiles.map((f) => fs.unlink(f)));
await fs.rmdir(tmpDir);
await fs.writeFile(
Why it matters: Usually legitimate, but worth confirming the paths can't be controlled by untrusted input.
Fix: Confirm which files are written/deleted and that paths cannot be influenced by untrusted input.
package.json has a 'prepare' script (runs on git/local installs and before publishing).
"prepare": "husky && pnpm build",
Why it matters: Usually a build step, but confirm it doesn't fetch or run remote code.
Fix: Usually a legitimate build step; confirm it only builds and does not fetch or execute remote code.
The component makes outbound network requests.
import http from "node:http";
Why it matters: Usually legitimate, but confirm the destinations are expected and no sensitive data leaves.
Fix: Confirm the destination hosts are expected and that no sensitive data is sent off-host.
An MCP tool surface (manifest or tool definitions) was found.
"mcpServers": {const server = new McpServer({await server.connect(new StdioServerTransport());
const transport = new StreamableHTTPServerTransport({Why it matters: Just context — review which tools it offers and their permissions.
Fix: Review the declared MCP tools and their permissions.
How attackers abuse these capabilities
Interactive labs on the attack class behind the rules above. They show the technique, not anything found in Browserbase MCP server.
Maintain Browserbase MCP server?
The free Claude Code plugin scans each npm, PyPI or MCP package with this report's open-source engine before the install command runs. It runs on your machine and needs no account.
Check your own component
Run the same evidence-backed scan on any MCP server, agent skill, or package.
Scan your own componentHow we determine this: deterministic static analysis (regex + AST), evidence-anchored, no code execution. Methodology →