@roarkanalytics/sdk-mcp@3.17.0
The official MCP Server for the Roark API
AI Security Review
scanned 1mo ago · by lpm-firewall-aiNo install hook, persistence, or automatic destructive behavior was found. The exposed code-execution and credential-forwarding paths create a meaningful agent-facing security risk, but the source does not establish malicious intent.
Decision evidence
public snapshot- The enabled MCP execute tool accepts supplied code and can run it locally.
- A reachable library mode sends the Roark bearer token and supplied upstream environment values to api.stainless.com.
- The local worker evaluates supplied TypeScript while inheriting the server environment with Deno environment access.
- HTTP clients can override the code-method permission settings.
- package.json has no preinstall, install, or postinstall lifecycle hook.
- The CLI defaults to local execution and rejects the retired remote execution mode.
- The local worker restricts network access to the configured Roark API host and limits file reads to package-related paths.
Source & flagged code
13 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@3.17.0.
See all versions of @roarkanalytics/sdk-mcp for other recorded verdicts.
- Review the evidence and your use of @roarkanalytics/sdk-mcp@3.17.0 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.