@agentvault/openai-bridge@0.1.10
AgentVault OpenAI Codex bridge for encrypted owner messages and rooms
AI Security Review
scanned 7d ago · by lpm-firewall-aiThe package automatically sends model usage metrics and agent identifiers to the AgentVault API, but the inspected README does not disclose this reporting or identify an opt-out. The reported fields do not include prompts or replies, so this warrants a warning rather than a publish block.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
Automatically after an authenticated channel connects and replies are processed.
Impact
AgentVault can associate model usage and token counts with the connected agent; the inspected reports do not include message text.
Mechanism
An authenticated telemetry reporter periodically posts hub and agent metadata plus model usage metrics to the API telemetry endpoint.
AI rationale
The package automatically sends model usage metrics and agent identifiers to the AgentVault API, but the inspected README does not disclose this reporting or identify an opt-out. The reported fields do not include prompts or replies, so this warrants a warning rather than a publish block.
Evidence
dist/index.jsREADME.md
Network endpoints1
api.agentvault.chat
Decision evidence
public snapshotAI called this Suspicious at 91.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- After the AgentVault channel connects, the bundled client creates an authenticated telemetry reporter and starts automatic 30-second flushing.
- The reporter posts telemetry to the configured API's /api/v1/telemetry/ingest endpoint with hub and agent identity and version metadata.
- For Codex replies, the bridge reports model, provider, latency, and input/output token counts; the reported fields do not include message text.
- The manifest has a prepublishOnly build script, but no install lifecycle hook.
- The entrypoint calls main only when invoked directly, rather than when imported.
Evidence against
- The telemetry reports usage and service identifiers rather than prompts or replies.
- The package's advertised purpose is to bridge addressed AgentVault messages to a dedicated Codex login.
- The inspected README does not disclose runtime telemetry or describe a way to disable it.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetwork
Source & flagged code
9 flaggeddist/index.jsView file
104038await init_bridge();
L104039: import { spawn as spawn2 } from "node:child_process";
L104040: import { realpathSync } from "node:fs";
...
L104050: checkBinary: checkVersion,
L104051: output: (text) => process.stdout.write(text + "\n"),
L104052: command: (binary, args2, options = {}) => new Promise((resolve4, reject) => {
L104053: 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 · L104038104038Trigger-reachable chain: manifest.main -> dist/index.js
L104038: await init_bridge();
L104039: import { spawn as spawn2 } from "node:child_process";
L104040: import { realpathSync } from "node:fs";
...
L104050: checkBinary: checkVersion,
L104051: output: (text) => process.stdout.write(text + "\n"),
L104052: command: (binary, args2, options = {}) => new Promise((resolve4, reject) => {
L104053: 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 · L104038Findings
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@0.1.10.
See all versions of @agentvault/openai-bridge for other recorded verdicts.
- Review the evidence and your use of @agentvault/openai-bridge@0.1.10 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.