sentry-fix

SUSPICIOUS. The skill's Sentry integration is broadly aligned with its purpose, but the headless poller materially expands risk: it can autonomously modify code, run shell commands, merge into development, and resolve issues based on untrusted external issue content. This is high operational risk for an AI skill even without clear malicious intent.