oh-my-codex@0.21.41
Multi-agent orchestration layer for OpenAI Codex CLI
AI Security Review
scanned 5d ago · by lpm-firewall-aiThe evaluation helper silently routes OPENAI_API_KEY to api.layofflabs.com for matching models by default. Because the key belongs to OpenAI and the user is not asked to select or confirm this separate endpoint, this is credential exfiltration.
Decision evidence
public snapshot- The eval collector automatically selects the layofflabs provider for matching internal model names and uses OPENAI_API_KEY as its credential variable.
- The collector configures that provider with https://api.layofflabs.com/v1 and binds its env_key to OPENAI_API_KEY.
- The eval script includes astra-defaults in its default config list and passes the provider setting into the collector loop.
- The credential flow is in an evaluation helper that must be run explicitly; it is not triggered by package import or ordinary CLI startup.
- The postinstall hook is limited to global installs and hydrates the package’s native runtime.
- The hard-coded GitHub issue targets appear in smoke-test command probes, not automatic account actions.
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@0.21.41.
See all versions of oh-my-codex for other recorded verdicts.
- Avoid installing oh-my-codex@0.21.41. 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.