learn-anything

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly instructs the agent to "use" $AUTH_TOKEN as a Bearer token and $API_URL (likely accessible in the agent/system context), which requires embedding or transmitting secret credentials verbatim in generated code/requests (e.g., Authorization headers in fetch/curl), creating a direct secret-exfiltration risk.

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md Phase 1/“Gather Knowledge” workflow explicitly instructs the agent to fetch and read open web content (e.g., "URL / webpage | WebFetch to get the content" and "Use web search and deep research" plus use of data-scraper/super-extract), so the agent will ingest untrusted third-party pages and use their content to drive slide generation and subsequent actions.