oh-my-codex@0.21.40
Multi-agent orchestration layer for OpenAI Codex CLI
AI Security Review
scanned 5d ago · by lpm-firewall-aiThe optional evaluator automatically routes matching models through api.layofflabs.com and supplies OPENAI_API_KEY as the provider credential. Since the credential is sent outside its issuing service without explicit consent, this is credential exfiltration despite requiring the evaluation tool to be run.
Decision evidence
public snapshot- The evaluator automatically selects the layofflabs provider for model names matching its internal-model pattern and defaults its credential variable to OPENAI_API_KEY.
- The selected provider config sends Codex requests to api.layofflabs.com and names OPENAI_API_KEY as the provider credential.
- The bundled evaluation script passes its configs to collectMatrix, which invokes the collector.
- The credential-forwarding path is in an evaluation tool, not the package’s normal import or install path; it requires running the evaluator.
- The postinstall hook is limited to global installs and hydrates the package’s own native runtime.
Source & flagged code
12 flaggedInstall-time lifecycle script matches a deterministic static-gate block pattern.
package.jsonView on unpkgPackage defines install-time lifecycle scripts.
package.jsonView on unpkgFindings
Affected versions and remediation
This report applies to oh-my-codex@0.21.40.
See all versions of oh-my-codex for other recorded verdicts.
- Avoid installing oh-my-codex@0.21.40. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.