@roarkanalytics/sdk-mcp@4.0.1
The official MCP Server for the Roark API
AI Security Review
scanned 23d ago · by lpm-firewall-aiThe reachable library path forwards existing service credentials to an unrestricted configurable endpoint, meeting the credential-forwarding warning policy. No separate block-eligible attack behavior was established.
Decision evidence
public snapshot- code-tool.js selects an environment-configurable remote endpoint without restricting its destination.
- The remote handler reads the existing Roark bearer token from the environment or SDK client.
- The remote request sends merged client credentials in a POST header.
- server.js includes the execution tool by default and selects remote execution when library callers omit execution options.
- The CLI defaults to local execution and rejects the retired remote mode.
- package.json has no automatic installation lifecycle hooks.
- No concrete malicious intent or covert package-controlled recipient was established.
- Credential forwarding requires a remote execution tool call; the normal CLI uses local execution.
Source & flagged code
12 flagged · loading sourceSource file is highly similar to a previously finalized malicious package; route for source-aware review.
instructions.jsView on unpkgPackage source references dynamic require/import behavior.
instructions.jsView on unpkg · L7Findings
Affected versions and remediation
This report applies to @roarkanalytics/sdk-mcp@4.0.1.
See all versions of @roarkanalytics/sdk-mcp for other recorded verdicts.
- Review the evidence and your use of @roarkanalytics/sdk-mcp@4.0.1 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.