@addai/node@0.29.1
Daemon that pairs a machine with your +Ai account and runs Claude / Codex / Kimi / Gemini agents on its behalf. Reachable via Supabase from Vault, Entity Studio, or any other +Ai surface.
AI Security Review
scanned 14d ago · by lpm-firewall-aiThe runtime persistently suppresses Claude’s bypass-permissions confirmation prompt, which creates a concrete agent safety risk. The inspected install hook only repairs the package’s PTY helper; the remaining integration writes are scoped to session configuration or caller-requested content.
Decision evidence
public snapshot- Daemon startup persists Claude’s bypass-permissions prompt setting in the user’s configuration.
- The other inspected integration locations concern per-session agent files, temporary MCP configuration, or session skill installation.
- The postinstall hook calls the package’s PTY helper repair routine.
- The manifest describes a daemon that runs agents on the user’s behalf.
- No source evidence inspected showed credential collection, exfiltration, or remote payload execution.
- The PreCompact hook is written to a session project settings file and is gated on a memory flag.
Source & flagged code
12 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgPackage source references child process execution.
dist/command-runner.jsView on unpkg · L46Findings
Affected versions and remediation
This report applies to @addai/node@0.29.1.
See all versions of @addai/node for other recorded verdicts.
- Review the evidence and your use of @addai/node@0.29.1 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.