agent-squad

SUSPICIOUS. The purpose is coherent with multi-agent orchestration, but the skill’s operational footprint relies on an undocumented `agent-squad` wrapper that likely handles gateway and task-system credentials without a verifiable publisher/install path or clear API routing.

No direct evidence of covert malware behavior (exfiltration, cryptomining, reverse shells, or file damage) is present in this module. However, it has a meaningful security risk: it uses execSync with interpolated shell command strings and untrusted inputs (notably process.env.EDITOR and fields derived from CLI/local config such as schedule and identifiers). If an attacker can influence environment variables or modify local squad/agent config files or names, this provides a plausible command-injection path that can lead to arbitrary command execution. Recommend hardening (use execFile/spawn with argv arrays, validate/escape cron expressions and identifiers, and avoid executing raw EDITOR strings).