mcpspend-cost-tracking

Warn

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONPERSISTENCEDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads and installs packages from the NPM registry, specifically @mcpspend/proxy and @mcpspend/mcp-server.
  • [COMMAND_EXECUTION]: The recommended installation uses npx --yes, which downloads and executes the initialization script from a remote registry without prompting the user for confirmation of the package source.
  • [PRIVILEGE_ESCALATION]: The init command modifies sensitive application-specific configuration files across multiple AI clients (Claude Desktop, Cursor, Windsurf, and VS Code). These files control how the AI agent interacts with the local system and external tools.
  • [PERSISTENCE]: By modifying the mcpServers configuration within various IDEs and desktop clients, the skill ensures the proxy wrapper is executed automatically whenever the AI environment starts. This establishes a persistent interception mechanism for all Model Context Protocol (MCP) traffic.
  • [DYNAMIC_EXECUTION]: The skill employs a wrapper pattern where it modifies existing tool commands to run as sub-processes of the @mcpspend/proxy. This allows the proxy to dynamically intercept and process all inputs and outputs between the AI agent and its tools at runtime.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 16, 2026, 10:43 PM
Security Audit — agent-trust-hub — mcpspend-cost-tracking