The AI review rates the findings as likely false positive, but the risk score (64/100) still counts them.
Analysis record
- Analysed
- 7 months ago 23 April 2026
- Version
- v0.1.0
- Artifact
- SHA256 d1733e42…987e6d
- Detail
- Full list
Evidence ledger
Ranked by severity · findings with a source location link to the code viewer
YARA Rule Matches
4 rules| Severity | Rule | Hits | Files | Metadata |
|---|---|---|---|---|
| HIGH | postinstall network communication Network communication detected | 2 | README.mdbuild/index.js | Risky Plugins Authors FP 30% |
| HIGH | postinstall file manipulation File system manipulation detected | 1 | build/index.js | Risky Plugins Authors FP 20% |
| HIGH | UsingShellInterpreterWhenExecutingOSCommands 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. | 1 | build/index.js | FP 10% |
| HIGH | postinstall system command System command execution detected | 2 | README.mdbuild/index.js | Risky Plugins Authors FP 10% |
Finding Categories
YARA Rules Matched
4 rules(6 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 none; evidence quality moderate.
Tool Poisoning Analysis
The findings summary shows 0 tool-poisoning findings. This is the most critical indicator for MCP packages. There are no hidden AI manipulation directives, no suspicious tool descriptions with embedded instructions, and no evidence of tool result interception. The package defines legitimate documentation tools without any poisoning architecture.
Credential and Network Access
The findings show 0 secret findings and 0 network findings. There is no credential-access activity targeting sensitive paths like .ssh/, .aws/, .kube/config, or specific secret names (GITHUB_TOKEN, AWS_SECRET_ACCESS_KEY). Without credential harvesting combined with network exfiltration, there is no exfiltration architecture present. The package reads no credentials beyond standard configuration.
IOC Findings Analysis
All 19 IOC findings are clearly XIOC extractor false positives:
XIOC-DOMAIN-date.now- JavaScript function name misread as domainXIOC-DOMAIN-request.params.name- Property access chain misread as domainXIOC-DOMAIN-item.name- Property access chain misread as domainXIOC-DOMAIN-notes.mdandXIOC-DOMAIN-workflowdetails.md- File extensions misread as domainsXIOC-URL-https://github.com/ryanjoachim/mcp-rtfm.git- Legitimate GitHub repository reference from documentationXIOC-URL-https://github.com/modelcontextprotocol/inspector- Legitimate MCP project referenceXIOC-DOMAIN-img.shields.io- Legitimate badge service domain
These match the known false-positive patterns: property chains (b.call, h.next), file names with extensions, and legitimate infrastructure domains. None represent actual network destinations.
Malware Signatures
The 6 malware-signature findings are code-smell rules (severity=low) that fire on standard JavaScript patterns. With 10 code-smell findings total, these represent generic code-quality matches, not actual malware.
Strongest Counterargument
The 12 high-severity findings and 0 user count could suggest an unvetted package. However, severity is inflated by the XIOC false positives and code-smell rules. The user count of 0 indicates a new/unpublished package, but newness alone does not indicate malice. The absence of any tool-poisoning, credential theft, or network exfiltration findings outweighs these concerns.
Conclusion
This package shows no evidence of malicious behavior. The findings are entirely from XIOC extractor noise (property chains, file names, legitimate URLs) and code-smell rules. The package implements legitimate documentation functionality without credential harvesting, network exfiltration, or tool poisoning.
Key Reasons
- Zero tool-poisoning findings - no AI manipulation directives detected
- Zero credential-access findings - no secret harvesting architecture
- Zero network findings - no exfiltration or suspicious outbound connections
- All IOC findings are XIOC extractor false positives from property chains and file names
False Positive Considerations
- XIOC-DOMAIN findings from property access chains (request.params.name, item.name)
- XIOC-DOMAIN findings from file extensions (notes.md, workflowdetails.md)
- XIOC-URL findings from legitimate GitHub and documentation references
- Code-smell rules matching standard JavaScript patterns
Reviewed 2026-04-27; 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