registry / record

@modelcontextprotocol/server-brave-search

npm / sealed 2026-06-16 / No. 54f21beb

Does @modelcontextprotocol/server-brave-search MCP work in Claude Code and Cursor? Throne executed @modelcontextprotocol/server-brave-search mcp in a single-use Firecracker microVM and replayed both recorded client behaviours. Compatibility verdict: needs your API key. Is @modelcontextprotocol/server-brave-search mcp safe? Static security scan: 2 finding(s) to review under security ruleset v1. The full per-step results, security findings, scan date, and evidence hash are below.

re-scan this server gate it in your CI

> throne registry @modelcontextprotocol/server-brave-search sealed

Upstream notice: the maintainer has deprecated this packagePackage no longer supported. Contact Support at https://www.npmjs.com/support for more info.

This server runs. It just needs your API keythe server installs and starts, then exits asking for an API key, Error: BRAVE_API_KEY environment variable is required

receiptsealed evidence

scan id: 54f21beb856443d1b69a41cf4d07805c
target: @modelcontextprotocol/server-brave-search
sealed at: 2026-06-16 20:57:21Z
evidence hash: sha256:6dd429455ba92af370d1253e083c41bf22ac229fe5c040f6706856c733c7dc7f

01connectFAIL0ms

server never launched: server process exited with code 1 before the MCP handshake, stderr: Error: BRAVE_API_KEY environment variable is required

02discoverSKIPPED0ms

not run, server never launched

03validate_schemasSKIPPED0ms

not run, server never launched

04smoke_test_toolsSKIPPED0ms

not run, server never launched

05error_handlingSKIPPED0ms

not run, server never launched

06streamingSKIPPED0ms

not run, server never launched

07resource_lifecycleSKIPPED0ms

not run, server never launched

08concurrent_callsSKIPPED0ms

not run, server never launched

09reconnectSKIPPED0ms

not run, server never launched

01connectFAIL0ms

server never launched: server process exited with code 1 before the MCP handshake, stderr: Error: BRAVE_API_KEY environment variable is required

02discoverSKIPPED0ms

not run, server never launched

03validate_schemasSKIPPED0ms

not run, server never launched

04smoke_test_toolsSKIPPED0ms

not run, server never launched

05error_handlingSKIPPED0ms

not run, server never launched

06streamingSKIPPED0ms

not run, server never launched

07resource_lifecycleSKIPPED0ms

not run, server never launched

08concurrent_callsSKIPPED0ms

not run, server never launched

09reconnectSKIPPED0ms

not run, server never launched

chatgpt desktop calibratingemulation profile pending real-traffic captureCOMING SOON

SECURITY: REVIEW / 2 finding(s) / review recommended

LOWTHR-INSTALL-03 / Install-time script execution (npm lifecycle)

"prepare" runs on git-dependency installs and local dev (not registry installs): 'npm run build'

package/package.json

LOWTHR-NET-05 / Hardcoded outbound endpoints

1 non-local endpoint host(s) referenced in code, verify each is expected for this server's purpose: api.search.brave.com (package/dist/index.js:109)

package/dist/index.js:109

VERDICT: NEEDS YOUR API KEYSANDBOXED RUN, submitted server executed in a disposable microVM, compatibility not assessable: needs credentials: the server installs and starts, then exits asking for an API key, Error: BRAVE_API_KEY environment variable is required / security: review, 2 finding(s), 0 highsealed by THRONE / No. 54f21beb / 2026-06-16

scopeattestation tuple

target: @modelcontextprotocol/server-brave-search npm
engine: sandboxed
claude code: calibration: partial recorded=9,spec=0,assumed=9
cursor: calibration: partial recorded=11,spec=0,assumed=7
chatgpt desktop: calibration: unavailable recorded=0,spec=2,assumed=16
steps exercised: 18 of 18
test suite: v1.0.0
security ruleset: v1.0.0
sealed at: 2026-06-16 20:57:21Z
evidence hash: sha256:6dd429455ba92af370d1253e083c41bf22ac229fe5c040f6706856c733c7dc7f
valid until: 2026-09-14

executed in a disposable microVM, created for this scan and destroyed after it. nothing outlives a run.

maintainer of this server? challenge this record: hello@usethrone.dev. tell us what we got wrong and we re-run it in the open.

this page renders the stored record of a real run. nothing on it is asserted without the execution that proved it.