@factiii/runner@0.18.1
Factiii Runner, run Board AI agents on a machine you control. Pairs with the Factiii web/mobile clients over WebRTC.
AI Security Review
scanned 19d ago · by lpm-firewall-aiInspected source supports a first-party agent extension installation and automatic update capability, which the supplied policy classifies as warning-only. Scanner claims of malicious credential redirection and install abuse were not substantiated by the inspected activation and credential paths.
Decision evidence
public snapshot- The manifest has no npm installation lifecycle hook; prepublishOnly builds publishing artifacts.
- The bundled tmux save script writes session layouts to local backup files.
- Runner workspace setup installs the Factiii plugin at Claude user scope and adds it to Codex.
- Setup also updates installed plugins automatically from the locally rendered package plugin, without a separate plugin consent prompt.
- Deployment approval and execution create a workspace Claude skill link; the board MCP uses the runner pairing token with its configured server.
- The plugin status hook exits when no workspace Redis socket is found.
- No inspected source established covert credential theft or malicious installation behavior.
- Plugin files come from the shipped Factiii integration rather than an unrelated remote payload.
Source & flagged code
14 flagged · loading sourceSource appears to send environment or credential material to an external endpoint.
dist/cli.jsView on unpkg · L38A package entrypoint or install-time lifecycle script reaches a source file with blocking dangerous behavior.
dist/cli.jsView on unpkg · L38Findings
Affected versions and remediation
This report applies to @factiii/runner@0.18.1.
See all versions of @factiii/runner for other recorded verdicts.
- Review the evidence and your use of @factiii/runner@0.18.1 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.