mcpspend-cost-tracking

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This prompt repeatedly shows and instructs embedding API keys (e.g., --key, Authorization: Bearer, env vars, JSON configs) into CLI commands and config files, which would require an LLM to include secret values verbatim in generated output (high exfiltration risk).

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). These links point to a small/unverified project that instructs users to run code via npx (which executes remote packages), modify local MCP configs, and rely on a GitHub account that appears to be a personal/lab repo and an external remote server domain — all of which are moderate-to-high risk without independent verification.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). This skill repeatedly invokes npx to fetch and execute the @mcpspend/proxy package at runtime (e.g., "npx -y @mcpspend/proxy@latest" — https://www.npmjs.com/package/@mcpspend/proxy), which pulls and runs remote code as a required dependency (it also communicates with the remote API at https://api.mcpspend.com), so the external content directly executes at runtime.