ln-821-npm-upgrader

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs the agent to fetch and read external documentation and migration guides from public sources (e.g., WebFetch to https://raw.githubusercontent.com/... and MCP tools like mcp__context7__query-docs, mcp__Ref__ref_search_documentation, and WebSearch) as mandatory workflow steps, meaning untrusted third‑party content can be read and influence upgrade/migration actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs at runtime to fetch missing shared files from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path}, and those fetched documents (e.g., breaking_changes_patterns.md, shared coordinator contracts) are required and directly influence the skill's decision-making and instructions, so this is a runtime external dependency that can control behavior.