export-solution

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill captures a runtime token (and uses envUrl) from verify-alm-prerequisites.js and then embeds that token directly into subsequent CLI invocations (e.g., --token "{token}" in multiple node scripts and curl-like requests), which requires the LLM to handle and output secret values verbatim, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.85). Phase 2.5/2.5+ and Phase 1.5 fetch and ingest outsider-authored free text from Microsoft Learn at runtime (via microsoft_docs_fetch / microsoft_docs_search), which is then summarized and placed into the agent’s LLM context.