hostinger-api-mcp@1.55.1
MCP server for Hostinger API
AI Security Review
scanned 2mo ago · by lpm-firewall-aiNo install-time attack is present, but the bundled agent-facing workflow directs remote code retrieval, unpinned package execution, and authenticated resource actions. This warrants a warning rather than a publish block because activation is explicit and no concrete exfiltration or stealth persistence was found.
Decision evidence
public snapshot- An agent-facing skill instructs a coding agent to execute the supplied workflow and to fetch code with curl.
- The bootstrap runs an unpinned latest MCP package through npx and can launch its login command.
- The skill then instructs the agent to install a further skill and grants it an authenticated provisioning workflow.
- package.json has no preinstall, install, or postinstall hook.
- The bootstrap uses shell:false and its credential check is limited to Hostinger token and OAuth credential locations.
- The runtime network and file operations are Hostinger API and deployment features invoked through MCP tools.
Source & flagged code
4 flagged · loading sourcePackage source invokes a package manager install command at runtime.
build.jsView on unpkg · L18This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
Findings
Affected versions and remediation
This report applies to hostinger-api-mcp@1.55.1.
See all versions of hostinger-api-mcp for other recorded verdicts.
- Review the evidence and your use of hostinger-api-mcp@1.55.1 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.