@phnx-labs/agents-cli
One CLI for all your AI coding agents - versions, config, cloud dispatch, sessions, and teams (now with first-class Grok Build CLI support)
This report shows the evidence for @phnx-labs/agents-cli@1.22.109. 4 of 132 known versions are blocked or flagged. The newest version, 1.23.43, is allowed. See all versions.
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThis is an automatic install-hook persistence chain with credential-accessing runtime behavior. The package mutates the user's agent execution environment and starts a long-running service without explicit post-install consent.
Decision evidence
public snapshot- The package runs a postinstall script automatically.
- Postinstall starts or restarts an always-on daemon without a prompt.
- The daemon writes and enables a user systemd service for persistence.
- Runtime reads Codex authentication data and sends bearer-authenticated usage requests to Anthropic.
- Shell configuration edits are prompted or opt-in in one postinstall branch.
- The service and shims are named for this package rather than disguised as another product.
Source & flagged code
26 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgPackage source references child process execution.
dist/bootstrap.jsView on unpkg · L425Findings
Affected versions and remediation
This report applies to @phnx-labs/agents-cli@1.22.109.
The versions list below shows the verdict for each recorded version.
- Avoid installing @phnx-labs/agents-cli@1.22.109. 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.