ida-reverse

SUSPICIOUS. The skill’s capabilities mostly match its stated reverse-engineering purpose, but it is a high-impact offensive/security-analysis skill that can open, inspect, and patch binaries and run local helper scripts. Data flow stays on localhost and there is no clear credential harvesting or third-party proxying, which argues against malware. Main concerns are the unpinned GitHub-based install from a personal repo, autostarted local service management, and the disproportionate real-world power granted to an AI agent for cracking/patching workflows.

No clear, explicit malware payload is visible in the provided fragment (no obvious exfiltration, credential harvesting, backdoor network beacons, or destructive file operations). However, there are notable supply-chain and security concerns: it auto-bootstraps the MCP component using PowerShell with ExecutionPolicy Bypass (increasing risk if the bootstrap script or fetched dependency is compromised), and it launches the MCP supervisor with '--unsafe' explicitly enabling py_eval/py_exec_file style execution. The process is also started hidden and relies on dynamic discovery of executable/module paths. Overall: likely a legitimate IDA integration tool, but the security posture is high-risk by design and has meaningful supply-chain exposure.

This module primarily acts as a scheduled-task persistence installer for a local watchdog.ps1, using stealth features (hidden task and hidden PowerShell window) and weakening script execution controls (-ExecutionPolicy Bypass). It also establishes long-lived, frequent execution and restart-on-failure behavior, all of which increase supply-chain security risk. No direct exfiltration/credential theft/network activity is observable in the provided fragment; confirming or ruling out malware requires inspection of watchdog.ps1 and the processes it manages.