@agent-infra/mcp-client
Score-based assessment (medium risk, 44/100). An AI review is available in the AI review tab.
No individual score drivers were recorded for this analysis.
Analysis record
- Analysed
- 7 months ago 19 April 2026
- Version
- v1.2.29
- Artifact
- SHA256 6712a2b0…c77a25
- Detail
- Full list
Evidence ledger
Ranked by severity · findings with a source location link to the code viewer
YARA Rule Matches
5 rules| Severity | Rule | Hits | Files | Metadata |
|---|---|---|---|---|
| HIGH | postinstall network communication Network communication detected | 2 | dist/index.mjsdist/index.js | Risky Plugins Authors FP 30% |
| HIGH | postinstall file manipulation File system manipulation detected | 4 | README.mddist/index.d.tsdist/index.mjs +1 more | Risky Plugins Authors FP 20% |
| HIGH | postinstall system command System command execution detected | 5 | README.mdpackage.jsondist/index.mjs +2 more | Risky Plugins Authors FP 10% |
| HIGH | postinstall file download File download activity detected | 3 | README.mddist/index.jsdist/index.mjs | Risky Plugins Authors FP 30% |
| HIGH | credential env files Environment configuration file path detected | 2 | dist/index.jsdist/index.mjs | Risky Plugins Authors FP 10% |
Finding Categories
YARA Rules Matched
5 rules(16 hits)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 pup; evidence quality weak.
This package presents a high risk score driven entirely by 16 malware-signature matches and 1 network finding. However, automated heuristic scanners frequently misclassify legitimate code—particularly in complex environments like Electron or tooling infrastructure—as malware due to code structure similarities or fingerprinting collisions. The absence of tool-poisoning findings indicates the package does not attempt to manipulate the AI agent through hidden directives. While the low user count (0) suggests this is an unproven library, there is no direct evidence of credential harvesting or exfiltration. The high risk score likely reflects 'score inflation' from aggressive YARA rules rather than an active threat.
Key Reasons
- High risk score (83.47) driven entirely by generic malware-signature matches which are high false-positive candidates
- Zero tool-poisoning findings detected despite the package's nature as an MCP client
- Zero credential-access or secret findings, indicating no immediate data theft patterns
- Low user count (0) suggests unproven software but does not confirm malicious intent
False Positive Considerations
- YARA malware-signature rules (16 findings) frequently trigger on legitimate code patterns or library bundles
- Score inflation where generic signature matches disproportionately inflate the risk_score
Reviewed 2026-04-13; recommended action: suppress false positive; model confidence 85%.
Source Code Not Available
Source code is not available for this version of the extension.
About This Extension
Frequently Asked Questions
Similar Extensions
Related extensions from the same publisher or marketplace