skills/nanocoai/nanoclaw/add-rtk/Gen Agent Trust Hub

add-rtk

Fail

Audited by Gen Agent Trust Hub on Jul 16, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill uses the dangerous curl | sh pattern to download and execute an installation script directly from a remote GitHub repository (https://raw.githubusercontent.com/rtk-ai/rtk/refs/heads/master/install.sh).- [EXTERNAL_DOWNLOADS]: Downloads a binary from an unverified third-party repository (rtk-ai/rtk) that is not part of the trusted vendor list.- [COMMAND_EXECUTION]: Modifies the agent platform's internal SQLite database (data/v2.db) using a script (scripts/q.ts) to inject custom container mounts.- [COMMAND_EXECUTION]: Programmatically alters the agent's settings.json configuration file to inject a PreToolUse hook, which intercepts and proxies all bash tool calls through the external rtk binary.
Recommendations
  • HIGH: Downloads and executes remote code from: https://raw.githubusercontent.com/rtk-ai/rtk/refs/heads/master/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
CRITICAL
Analyzed
Jul 16, 2026, 04:18 PM
Security Audit — agent-trust-hub — add-rtk