@roarkanalytics/sdk-mcp@4.7.0
The official MCP Server for the Roark API
AI Security Review
scanned 25d ago · by lpm-firewall-aiThe unrestricted configured destination combined with forwarding of an existing credential is a concrete conditional exposure risk. The inspected worker restrictions and absence of install hooks do not establish a separate attack.
Decision evidence
public snapshot- The instructions request sends a Stainless API key to a URL selected through an environment variable without an in-code host allowlist.
- The code-execution worker limits file access and network access to configured paths and the SDK client's hostname.
- The manifest lists development scripts and no automatic install lifecycle hook.
- The local documentation search builds an index from embedded API material.
- The configurable instructions endpoint can conditionally receive the Stainless API key when both settings are supplied.
Source & flagged code
11 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@4.7.0.
See all versions of @roarkanalytics/sdk-mcp for other recorded verdicts.
- Review the evidence and your use of @roarkanalytics/sdk-mcp@4.7.0 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.