registry / @ddongule/i18n-mcp / 1.1.0

@ddongule/i18n-mcp@1.1.0

An MCP (Model Context Protocol) server that exposes i18n automation as **AI-callable tools**.

AI Security Review

scanned 8h ago · by lpm-firewall-ai

The create_i18n_pr MCP tool has command injection through its branch argument. No install-time execution is present.

Static reason

No blocking static signals were detected.

Trigger

An MCP client invokes create_i18n_pr with a crafted branch value.

Impact

Arbitrary commands execute with the MCP server process's permissions.

Mechanism

Unsanitized shell interpolation in execSync.

Rationale

Source inspection confirms a reachable shell-injection flaw in an MCP tool. The package has no lifecycle hooks or automatic hostile behavior, so it should be warned rather than blocked as malicious.

Evidence

package.jsondist/index.jslocalesgoogle-credentials.json

Decision evidence

public snapshot

AI called this Suspicious at 94.0% confidence as Critical Vulnerability with low false-positive risk.

Evidence for warning

dist/index.js:693 interpolates MCP-controlled branch into execSync shell command.
A crafted branch value can append shell commands before git executes.
create_i18n_pr is a remotely invokable MCP tool, enabling host command execution.

@ddongule/i18n-mcp@1.1.0

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code