Audit History

Forty-three static alerts are false positives caused by Markdown backticks, code fences, placeholder media URLs, or ordinary documentation links. The unpinned npx bootstrap command is a confirmed supply-chain risk, and the skill also enables consequential X account mutations without confirmation safeguards.

Most static external-command findings are Markdown backticks or app identifiers, not Ruby execution. The skill still contains real infsh and npx command examples that can post, delete, message, follow, install related skills, and contact inference.sh services.

Most static external-command findings are Markdown backticks or app identifiers, not Ruby execution. The skill still contains real infsh and npx command examples that can post, delete, message, follow, install related skills, and contact inference.sh services.

Static command-execution findings are mostly true positives because the skill intentionally instructs agents to run the external infsh and npx CLIs. No confirmed malicious intent, credential exfiltration, or prompt injection was found, but the skill can perform high-impact Twitter/X account actions and should be published only with clear user authorization warnings.

Static analyzer detected 37 external command patterns and 6 network URLs, but all are false positives. The findings are documentation examples in SKILL.md showing CLI usage, not executable code. The skill uses allowed-tools restriction (Bash(infsh *)) limiting execution to the inference.sh CLI only. Weak crypto findings at lines 3 and 27 are false positives matching description text. Low risk due to external service dependency on inference.sh platform.