crypto-com-mcp-skill
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is configured to interact exclusively with official Crypto.com MCP endpoints (e.g.,
https://mcp.crypto.com/market-data/mcp) for fetching public market data. - [SAFE]: Explicit security guardrails are implemented within the instructions to ensure the agent remains in a read-only state, specifically prohibiting trading, account modifications, or access to private data.
- [SAFE]: The included validation script (
scripts/validate.sh) enforces best practices by verifying the presence of safety markers, correct endpoints, and help-first discovery patterns. - [SAFE]: No instances of prompt injection, data exfiltration, or obfuscation were identified. Dependencies are limited to a pre-installed helper tool (
uxc) and standard system utilities.
Audit Metadata