skill-generator

SUSPICIOUS: the core scaffolding behavior is consistent with a skill generator, but the footprint includes an under-specified external `maestro` CLI and external search-derived context feeding a write-capable workflow. This is not confirmed malware, but it has medium risk from transitive trust and prompt-injection exposure.

No direct evidence of embedded malware (no explicit credential theft, backdoors, or exfiltration endpoints) is visible in this fragment. However, it creates a significant security risk by embedding untrusted prompt/content (from local files and runtime context) into a shell command executed via Bash with incomplete escaping, and by writing outputs to config-driven paths without visible validation. Treat as a potentially dangerous orchestration layer that requires hardening (argv-safe process spawning, robust escaping, and strict path allowlisting/jails) before use in attacker-influenced scenarios.

The code scaffolds skill directories and writes a SKILL.md using values from skill-config.json. It contains no direct indicators of malware (no networking/backdoors/obfuscation), but it uses unsafe practices: unvalidated interpolation of configuration into shell commands and filesystem paths, and generation of documentation that contains executable snippets referencing attacker-controlled paths. These behaviors enable path traversal, file overwrite, and command-injection vectors if the configuration or downstream execution context is attacker-controlled. Mitigation: validate and sanitize inputs, avoid shell interpolation, use safe filesystem APIs, and do not embed runnable commands containing untrusted data.