mcp-developer

Fail

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: CRITICALEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The documentation URL https://jeffallan.github.io/claude-skills/skills/api-architecture/mcp-developer/ is blacklisted by automated security scanners.
  • [REMOTE_CODE_EXECUTION]: The skill provides instructions to execute code from external registries via npx @modelcontextprotocol/create-server and npx @modelcontextprotocol/inspector.
  • [COMMAND_EXECUTION]: The implementation workflow requires executing shell commands including pip install mcp and Node.js package execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill builds Model Context Protocol (MCP) servers that ingest data from external sources including APIs, databases, local file systems, and Git repositories. Evidence: 1. Ingestion points: Found in references/resources.md (File, DB, API, Git) and references/tools.md (DB, File, HTTP API). 2. Boundary markers: Mostly absent in provided code snippets. 3. Capability inventory: Includes fs.readFile, db.query, httpx.get, and shell command execution. 4. Sanitization: Minimal validation provided in examples (regex for table names, path resolution for files). These ingestion points represent an attack surface where untrusted data can influence agent behavior.
Recommendations
  • CRITICAL: 1 file(s) identified as malware by FileRep - DO NOT USE
  • AI detected serious security threats
  • Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Sep 5, 2026, 11:38 AM
Security Audit — agent-trust-hub — mcp-developer