registry / @agentlayer.tech/wallet / 0.1.108
@agentlayer.tech/wallet@0.1.108
Universal AgentLayer wallet installer for OpenClaw, Codex, Claude Code, and Hermes.
AI Security Review
scanned 13d ago · by lpm-firewall-aiThe launcher is user-invoked, has no lifecycle hook, and its agent configuration writes occur through explicit install commands. The flagged subprocess behavior is confined to a temporary smoke-test script or bounded local daemon management; the inspected source did not establish a concrete attack.
Static reason
One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source
Trigger
The user runs the wallet CLI; no package install lifecycle hook is declared.
Impact
No confirmed malicious effect or credential exposure was established from the inspected source.
Mechanism
The CLI records limited command metadata to the package gateway and performs agent configuration updates through explicit install commands.
AI rationale
The launcher is user-invoked, has no lifecycle hook, and its agent configuration writes occur through explicit install commands. The flagged subprocess behavior is confined to a temporary smoke-test script or bounded local daemon management; the inspected source did not establish a concrete attack.
Evidence
package.jsonbin/openclaw-agent-wallet.mjsbin/lib/evm-daemon.mjsconnectors/scripts/smoke-packed-packages.mjs
Network endpoints2
agent-layer-production.up.railway.app127.0.0.1
Decision evidence
public snapshotAI called this Clean at 88.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- package.json exposes a user-run wallet CLI and contains no lifecycle hooks.
- CLI telemetry records limited install and command metadata, with a disable flag, and flushes it to the configured package gateway.
- The indicated Claude cache location updates MCP configuration only through the CLI’s explicit install flow.
- The daemon helper limits takeover to loopback services and verifies process ownership before signaling.
- The connector smoke script operates in a temporary test directory and installs its test dependencies with scripts disabled.
Evidence against
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
NoLicense
Source & flagged code
6 flaggedconnectors/scripts/smoke-packed-packages.mjsView file
1import { execFileSync } from "node:child_process";
L2: import { createHash } from "node:crypto";
High
Child Process
Package source references child process execution.
connectors/scripts/smoke-packed-packages.mjsView on unpkg · L1bin/openclaw-agent-wallet.mjsView file
•matchType = previous_version_dangerous_delta
matchedPackage = @agentlayer.tech/wallet@0.1.89
matchedIdentity = npm:QGFnZW50bGF5ZXIudGVjaC93YWxsZXQ:0.1.89
similarity = 0.684
summary = stored previous version shares package body but lacks this dangerous source file
Critical
Previous Version Dangerous Delta
This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
bin/openclaw-agent-wallet.mjsView on unpkgFindings
1 Critical4 High4 Medium5 Low
CriticalPrevious Version Dangerous Deltabin/openclaw-agent-wallet.mjs
HighChild Processconnectors/scripts/smoke-packed-packages.mjs
HighShellbin/openclaw-agent-wallet.mjs
HighSame File Env Network Executionbin/lib/evm-daemon.mjs
HighRuntime Package Installbin/openclaw-agent-wallet.mjs
MediumNetwork
MediumEnvironment Vars
MediumShips Build Helpersetup.sh
MediumStructural Risk Force Deep Review
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
LowNo License
Affected versions and remediation
This report applies to @agentlayer.tech/wallet@0.1.108.
See all versions of @agentlayer.tech/wallet for other recorded verdicts.
- Review the evidence and your use of @agentlayer.tech/wallet@0.1.108 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.