debug-fe-be-integration

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly instructs the agent to read terminal logs and source files and to "show the actual code from Grep results" and also calls services with project IDs/placeholders, which can cause the agent to include secret values (API keys, tokens, credentials) verbatim in its output — an exfiltration risk.

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

• Third-party content exposure detected (medium risk: 0.30). Step 1 and Step 2 use external web/search and issue-reading at runtime (e.g., Firecrawl web search and Context7 library resolution, plus Sentry issue search), so outsider-authored free text can be ingested without first selecting a specific trusted item.