registry / @agentvault/openai-bridge / 0.1.4
@agentvault/openai-bridge@0.1.4
AgentVault OpenAI Codex bridge for encrypted owner messages and rooms
AI Security Review
scanned 14d ago · by lpm-firewall-aiThe manifest has no install-time hook; its prepublishOnly script only builds the package. Inspected runtime code showed a constrained, separate Codex environment and no confirmed malicious behavior.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
A user invokes the bridge CLI or starts its configured AgentVault channel.
Impact
No confirmed malicious effect was established from the inspected source.
Mechanism
The bridge runs Codex in its isolated state layout and validates the resulting reply.
AI rationale
The manifest has no install-time hook; its prepublishOnly script only builds the package. Inspected runtime code showed a constrained, separate Codex environment and no confirmed malicious behavior.
Decision evidence
public snapshotAI called this Clean at 91.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- package.json declares prepublishOnly as a build command and ws as a runtime dependency.
- The inspected ws runtime code formats the standard WebSocket Origin header.
- The prepared ws package exposes its normal Node entrypoints and browser fallback.
- dist/index.js launches Codex with a read-only sandbox and a dedicated environment rooted in the bridge state directory.
- The bridge verifies the Codex final reply before returning it.
Evidence against
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShellWebSocket
HighEntropyStringsUrlStrings
Oversized source lightweight scan
Source & flagged code
9 flaggeddist/index.jsView file
102463await init_bridge();
L102464: import { spawn as spawn2 } from "node:child_process";
L102465: import { realpathSync } from "node:fs";
...
L102475: checkBinary: checkVersion,
L102476: output: (text) => process.stdout.write(text + "\n"),
L102477: command: (binary, args2, options = {}) => new Promise((resolve4, reject) => {
L102478: const child = spawn2(binary, args2, { env: options.env ?? { PATH: "/usr/bin:/bin", HOME: process.env.HOME, XDG_RUNTIME_DIR: process.env.XDG_RUNTIME_DIR, DBUS_SESSION_BUS_ADDRESS: p...
Critical
Command Output Exfiltration
Source executes local commands and sends command output to an external endpoint.
dist/index.jsView on unpkg · L102463102463Trigger-reachable chain: manifest.main -> dist/index.js
L102463: await init_bridge();
L102464: import { spawn as spawn2 } from "node:child_process";
L102465: import { realpathSync } from "node:fs";
...
L102475: checkBinary: checkVersion,
L102476: output: (text) => process.stdout.write(text + "\n"),
L102477: command: (binary, args2, options = {}) => new Promise((resolve4, reject) => {
L102478: const child = spawn2(binary, args2, { env: options.env ?? { PATH: "/usr/bin:/bin", HOME: process.env.HOME, XDG_RUNTIME_DIR: process.env.XDG_RUNTIME_DIR, DBUS_SESSION_BUS_ADDRESS: p...
Critical
Trigger Reachable Dangerous Capability
A package entrypoint or install-time lifecycle script reaches a source file with blocking dangerous behavior.
dist/index.jsView on unpkg · L102463Findings
2 Critical6 High5 Medium5 Low
CriticalCommand Output Exfiltrationdist/index.js
CriticalTrigger Reachable Dangerous Capabilitydist/index.js
HighChild Processdist/index.js
HighShell
HighSame File Env Network Executiondist/index.js
HighCredential Redirect Persistencedist/index.js
HighTrigger Reachable Command Output Exfiltrationdist/index.js
HighSemantic Analysis Limiteddist/index.js
MediumNetwork
MediumEnvironment Vars
MediumOversized Source Filedist/index.js
MediumOversized Cli Entrypointdist/index.js
MediumStructural Risk Force Deep Review
Affected versions and remediation
This report applies to @agentvault/openai-bridge@0.1.4.
See all versions of @agentvault/openai-bridge for other recorded verdicts.
- Review the evidence and your use of @agentvault/openai-bridge@0.1.4 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.