MCP Registry

@twilio-alpha/openapi-mcp-server

00f73f00-a021-5828-a699-8cff03115597 | Store v0.7.0
64/ 100
MEDIUM risk
AI review verdict
Review before use

The AI review rates the findings as likely false positive, but the risk score (64/100) still counts them.

Analysis record

Analysed
8 months ago 10 March 2026
Version
v0.7.0
Detail
Full list

Evidence ledger

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

29 findings

YARA Rule Matches

7 rules
SeverityRuleHitsFilesMetadata
HIGHpostinstall file download

File download activity detected

9
build/server.jspackage.jsonbuild/server.d.ts +6 more
Risky Plugins Authors FP 30%
HIGHpostinstall obfuscation

Code obfuscation techniques detected

1
build/utils/http.js
Risky Plugins Authors FP 20%
HIGHpostinstall crypto operations

Cryptographic operations detected

1
build/utils/logger.js
Risky Plugins Authors FP 30%
HIGHpostinstall network communication

Network communication detected

1
build/server.js
Risky Plugins Authors FP 30%
HIGHpostinstall file manipulation

File system manipulation detected

6
build/utils/http.jsbuild/types.d.tsbuild/utils/loadTools.js +3 more
Risky Plugins Authors FP 20%
HIGHpostinstall environment access

Environment variable access detected

3
build/types.d.tsbuild/utils/readSpecs.d.tsbuild/utils/args.d.ts
Risky Plugins Authors FP 40%
HIGHpostinstall system command

System command execution detected

1
README.md
Risky Plugins Authors FP 10%

Finding Categories

22
Malware Signatures
1
Secrets
1
Network
29
IoC Indicators

YARA Rules Matched

7 rules(22 hits)
postinstall file download postinstall obfuscation postinstall crypto operations postinstall network communication postinstall file manipulation postinstall environment access postinstall system command

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 weak.

This package, an OpenAPI specification server from a recognized developer (twilio-alpha), demonstrates a classic high-risk-score/low-risk-reality profile common in enterprise SDKs.

Tool Poisoning Assessment: There are 0 tool poisoning findings. This is the strongest safety signal for an MCP server. The package exposes API specs without attempting to manipulate the AI agent's behavior through hidden directives.

Credential & Network Scope: No network findings (NET-FETCH/NET-SOCKET_IO) and no secret/credential findings (SECRET). The lack of network signatures is significant because it eliminates the most common MCP attack vector (credential exfiltration). The package reads process.env for configuration (standard Node.js behavior), but there is no evidence of harvesting or exfiltrating specific secrets like SSH keys or AWS tokens.

Finding Drivers: The "CRITICAL" risk score (85.28) is driven by 58 generic YARA malware signatures and 36 IoCs. Given the lack of specific threat evidence and the benign purpose of the tool, these are almost certainly false positives from bundled node_modules or minified code patterns triggering legacy antivirus heuristics. The top_findings field is null, which supports the theory that these flags lack specific context.

Conclusion: This appears to be a legitimate tool for exposing OpenAPI specs to AI agents, inflated by noisy generic rules.

Key Reasons

  • Zero tool poisoning findings (the primary MCP threat is absent)
  • Zero network findings eliminates the exfiltration vector
  • Zero secret/credential findings despite high risk score
  • Publisher is 'twilio-alpha-ci' (consistent with package name), not typosquatting

False Positive Considerations

  • YARA malware-signature rules (22 findings) matching generic JS patterns in node_modules
  • XIOC extractor noise (36 findings) from bundled dependencies and minified code
  • Risk score inflation from legacy antivirus heuristics applied to modern SDK code

Reviewed 2026-04-12; recommended action: suppress false positive; model confidence 75%.

Source Code Viewer

Source Code Not Available

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

About This Extension

A Model Context Protocol server that to expose OpenAPI specs.

Frequently Asked Questions