Skills
skills/nansen-ai/nansen-cli/nansen-mpp-payment/Gen Agent Trust Hub

nansen-mpp-payment

Fail

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides installation instructions (curl -fsSL https://tempo.xyz/install | bash) that pipe a remote script directly into a shell. This provides the host of the script with unverified execution capabilities on the user's system, a high-risk pattern for supply chain attacks.
  • [EXTERNAL_DOWNLOADS]: The skill requires downloading the tempo CLI from the external domain tempo.xyz.
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute tempo and nansen CLI commands for wallet management and network operations.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data from the Nansen API (api.nansen.ai) while having the capability to execute shell commands via the tempo tool. It lacks defined boundary markers or sanitization logic (file SKILL.md) to handle potentially adversarial content within API responses, which could lead to unauthorized command execution if the API returns malicious instructions.
Recommendations
  • HIGH: Downloads and executes remote code from: https://tempo.xyz/install - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
CRITICAL
Analyzed
Apr 30, 2026, 04:22 AM