sentry

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime path calls the Sentry API (e.g., scripts/sentry_api.py → request_json() → urlopen() → json.loads(body)), and the returned issue/event JSON is outsider-authored production error content that the script prints into the agent’s LLM context (including stack traces/entries when enabled).