@makerx/yoke@0.9.0
Lean, user-invoked Claude Code skills for MakerX AI-assisted delivery, plus the CLI they call.
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe flagged credential and process flows belong to the package's review CLI: provider credentials are directed to the configured provider base URL, and the CLI launches the selected review harness when invoked. Skill writes and CI setup are explicit commands; the manifest declares no install hook.
Decision evidence
public snapshot- The manifest exposes a CLI and has only a prepublishOnly build hook; no install hook is declared.
- Provider endpoints are derived from the configured base URL, and provider credentials are passed to that endpoint.
- The CLI launches Claude or Codex as a review harness; review starts from an explicit CLI action.
- Writing skills under .claude/skills is performed by the explicit sync command.
- The flagged Azure and Microsoft URLs are used for GitHub OIDC and Entra token exchange, not as a fixed credential collection gateway.
- The source does not establish automatic command-output exfiltration or agent configuration mutation.
Source & flagged code
5 flaggedManifest-reachable source sends a prompted API credential to a fixed unofficial gateway and persists the redirection.
dist/cli.jsView on unpkgFindings
Affected versions and remediation
This report applies to @makerx/yoke@0.9.0.
See all versions of @makerx/yoke for other recorded verdicts.
- Review the evidence and your use of @makerx/yoke@0.9.0 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.