registry / @agentvault/agentvault / 0.23.54
@agentvault/agentvault@0.23.54
⚠ Under reviewEnd-to-end encrypted messaging for AI agents: connects an agent to its owner, rooms and other agents on AgentVault.
AI Security Review
scanned 9d ago · by lpm-firewall-aiThe inspected credential requests use configured AgentVault API paths, so the available evidence does not establish exfiltration or another concrete attack. The large bundled runtime sources and incomplete analysis leave behavior unresolved, so manual review is appropriate.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
Importing the library or running the CLI or extension could activate runtime paths that were not fully resolved in this inspection.
Impact
The inspected requests do not show an unrelated recipient or theft; unresolved runtime behavior prevents a complete safety determination.
Mechanism
The inspected code makes requests to API paths formed from account configuration and sends device or API credentials in authorization headers.
AI rationale
The inspected credential requests use configured AgentVault API paths, so the available evidence does not establish exfiltration or another concrete attack. The large bundled runtime sources and incomplete analysis leave behavior unresolved, so manual review is appropriate.
Evidence
package.jsondist/index.jsdist/cli.jsdist/openclaw-entry.js
Network endpoints1
api.agentvault.chat
Decision evidence
public snapshotAI called this Manual Review at 68.0% confidence as Unknown with medium false-positive risk.
AI review observations
- The package exposes a library, CLI, and OpenClaw extension; its prepublishOnly hook runs the build.
- The account resolver defaults the API URL to https://api.agentvault.chat and permits account configuration to override it.
- Inspected message polling and hub skill requests send device or API credentials to URLs built from the configured API URL.
- The available source is readable, but the bundled CLI and library are large and analysis coverage is incomplete; the inspected excerpts do not resolve all runtime behavior.
Evidence against
- The inspected credential requests target configured AgentVault API paths; those excerpts do not establish a package-controlled unrelated recipient or credential theft.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsPossibleObfuscation
Source & flagged code
10 flaggeddist/cli.jsView file
129Source sends the broad process environment to a literal external destination.
L129: // ../../node_modules/libsodium-sumo/dist/modules-sumo-esm/libsodium-sumo.mjs
L130: var __filename, __dirname, url, path, Module, Module, root, window_, crypto_, randomValuesStandard, crypto2, randomValueNodeJS, _Module, libsodium_sumo_default;
L131: var init_libsodium_sumo = __esm({
...
L224: Module2.useBackupModule = function() {
L225: return new Promise(function(resolve6, reject2) {
L226: var Module3 = {};
...
L277: quit_2 = (status, toThrow) => {
L278: process.exitCode = status;
L279: throw toThrow;
...
L291: xhr.responseType = "arraybuffer";
L292: xhr.send(null);
L293: return new Uint8Array(xhr.response);
Critical
Hardcoded Runtime Data Exfiltration
Source sends credentials or rich application records to a package-controlled external receiver enabled by default.
dist/cli.jsView on unpkg · L12993502});
L93503: import { execSync } from "node:child_process";
L93504: import { existsSync as existsSync2, readFileSync as readFileSync3, writeFileSync, chmodSync, readdirSync, statSync } from "node:fs";
High
93615try {
L93616: execSync("curl -sf --max-time 3 http://127.0.0.1:18789/", { stdio: "pipe" });
L93617: return { name: "Gateway port (18789)", status: "pass", message: "Responding" };
...
L93629: }
L93630: const home = process.env.HOME ?? "";
L93631: const warnings = [];
Findings
1 Critical7 High6 Medium5 Low
CriticalHardcoded Runtime Data Exfiltrationdist/cli.js
HighChild Processdist/cli.js
HighShell
HighSame File Env Network Executiondist/cli.js
HighCredential Redirect Persistencedist/index.js
HighCommand Output Exfiltrationdist/cli.js
HighCross File Remote Execution Contextdist/cli.js
HighSemantic Analysis Limiteddist/cli.js
MediumDynamic Requiredist/openclaw-entry.js
MediumNetwork
MediumEnvironment Vars
MediumOversized Source Filedist/index.js
MediumOversized Cli Entrypointdist/cli.js
Affected versions and remediation
This report applies to @agentvault/agentvault@0.23.54.
See all versions of @agentvault/agentvault for other recorded verdicts.
- Review the evidence and your use of @agentvault/agentvault@0.23.54 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.