Coding Agent Loops
Security checks across malware telemetry and agentic risk
Overview
This is a transparent markdown guide for running coding agents in persistent tmux retry loops, with disclosed operational risks but no hidden payload or deceptive behavior found.
Before installing, confirm you trust ralphy-cli and any coding-agent credentials used by the workflow. Run loops in a branch or worktree, avoid full-auto or skip-test modes for sensitive repositories, monitor active tmux sessions, kill them when finished, and review git diff, logs, and test results before accepting changes.
SkillSpector
By NVIDIA
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
VirusTotal
66/66 vendors flagged this skill as clean.