sentry-observability

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill's CLI scripts (e.g., scripts/list_issues.sh, scripts/issue_detail.sh, scripts/triage_score.sh and the shared api_call in scripts/_common.sh) call the Sentry API (https://sentry.io/api/0/), fetch and parse issues, events, breadcrumbs and other event text coming from Sentry (user-generated/untrusted data), which the agent is expected to read and display as part of its workflow—exposing it to indirect prompt injection risk.