MCP Registry

@agent-infra/mcp-server-commands

92734b51-88d4-51d3-a0e1-829142d4a54c | Store v1.2.29
43/ 100
MEDIUM risk
AI review verdict
Needs follow up

From the RiskyPlugins AI security review of the observed evidence.

Analysis record

Analysed
7 months ago 18 April 2026
Version
v1.2.29
Artifact
SHA256 17e614e2…0298f5
Detail
Full list

Evidence ledger

Ranked by severity · findings with a source location link to the code viewer

40 findings

YARA Rule Matches

7 rules
SeverityRuleHitsFilesMetadata
HIGHpostinstall network communication

Network communication detected

3
README.mddist/index.cjsdist/index.js
Risky Plugins Authors FP 30%
HIGHpostinstall file manipulation

File system manipulation detected

5
package.jsondist/index.cjsdist/server.cjs +2 more
Risky Plugins Authors FP 20%
HIGHUsingShellInterpreterWhenExecutingOSCommands

Arbitrary OS command injection vulnerabilities are more likely when a shell is spawned rather than a new process, indeed shell meta-chars can be used (when parameters are user-controlled for instance) to inject OS commands. For more information checkout the CWE-78 (https://cwe.mitre.org/data/definitions/78.html) advisory.

3
dist/index.cjsdist/index.jsdist/server.js
FP 10%
HIGHpostinstall system command

System command execution detected

7
README.mddist/index.cjsLICENSE +4 more
Risky Plugins Authors FP 10%
HIGHpostinstall file download

File download activity detected

1
README.md
Risky Plugins Authors FP 30%
HIGHpostinstall obfuscation

Code obfuscation techniques detected

4
dist/index.cjsdist/server.cjsdist/index.js +1 more
Risky Plugins Authors FP 20%
HIGHcredential env files

Environment configuration file path detected

2
dist/index.cjsdist/index.js
Risky Plugins Authors FP 10%

Network Indicators

Concrete URLs, domains, IPs, emails, and hashes extracted from the analyzed artifact.

77 total

Loading network indicators

Long lists can take a few seconds.

Finding Categories

25
Malware Signatures
77
IoC Indicators

YARA Rules Matched

7 rules(25 hits)
postinstall network communication postinstall file manipulation UsingShellInterpreterWhenExecutingOSCommands postinstall system command postinstall file download postinstall obfuscation credential env files

MCP Server Analysis

MCP servers expose tools and resources to AI assistants. Unlike browser extensions, they run as standalone processes with direct system access. Tool definitions are analyzed for prompt injection, data exfiltration, and tool poisoning patterns.

AI Security Report

AI Security Review

Evidence context: threat category unknown malware; evidence quality weak.

This package presents a high-risk signal (Risk Score 100) based on artifact scanning, but lacks the behavioral evidence required to confirm malicious activity. The verdict is cautious due to the package's explicit capability: running arbitrary commands is a high-severity feature, yet the scan found zero tool-poisoning attempts and zero network activity (C2/exfiltration).

The 87 IoCs and 25 malware signatures are characteristic of 'Score Inflation' from bundled dependencies in node_modules or dist/ folders—common in MCP packaging. Without specific behavioral telemetry (e.g., detected outbound connections to unknown domains, hidden prompt injection in tool descriptions), we cannot differentiate between a vulnerable/unoptimized package and an active threat. The package name suggests it is an 'agent-infra' utility, which may legitimately bundle heavier dependencies. While the 'run arbitrary commands' functionality requires strict runtime scrutiny, the current evidence bundle contains only static detection noise (IoCs/YARA) and no concrete proof of misuse.

Key Reasons

  • No network findings (NET-FETCH, NET-SOCKET_IO) to support exfiltration theories.
  • 0 tool-poisoning findings despite explicit command execution capabilities.
  • 87 IoCs are likely false positives from bundled dependencies rather than active C2 infrastructure.
  • Threat score 100 is statistically characteristic of dependency noise, not confirmed malware.

False Positive Considerations

  • IoCExtractor: Matched generic domains/CDNs/IPv6 fragments in bundled dependencies.
  • MalwareSignature: YARA/Code-smell rules firing on minified or bundled utility code.

Reviewed 2026-04-13; recommended action: reanalyze; model confidence 55%.

Source Code Viewer

Source Code Not Available

Source code is not available for this version of the extension.

About This Extension

An MCP server to run arbitrary commands

Frequently Asked Questions