@roarkanalytics/sdk-mcp@4.32.1
The official MCP Server for the Roark API
AI Security Review
scanned 23h ago · by lpm-firewall-aiThe local execution feature grants submitted code broad environment access, presenting a real conditional exposure risk. The behavior is caller-invoked and no independent attack behavior was verified, so this warrants a warning rather than a block.
Decision evidence
public snapshot- The package exposes a code execution tool by default.
- The local worker runs caller-supplied code and returns its result and logs.
- The Deno worker receives broad environment-variable access, creating a risk that submitted code could expose environment values through its output.
- The sandbox restricts network access to the Roark API host.
- The code tool is an explicit runtime feature, and the inspected source shows no automatic installation-time attack or unrelated credential transfer.
- The Stainless-hosted execution mode is rejected as retired; local execution is the supported mode.
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 unpkgFindings
Affected versions and remediation
This report applies to @roarkanalytics/sdk-mcp@4.32.1.
See all versions of @roarkanalytics/sdk-mcp for other recorded verdicts.
- Review the evidence and your use of @roarkanalytics/sdk-mcp@4.32.1 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.