sage

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow uses the c7 MCP server for official library documentation and falls back to g MCP tools (e.g., mcp__g__search_and_scrape / scrape_page) which fetch and extract arbitrary third-party web page text into the LLM context, creating an outsider free-text injection surface.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill uses the g MCP tools (mcp__g__search_and_scrape and mcp__g__scrape_page) at runtime to fetch arbitrary external web URLs (i.e., http(s)://... pages returned/scraped by mcp__g__scrape_page), and that fetched content is injected into the agent's context and directly controls its responses — so external scraped URLs are a runtime dependency that controls prompts.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)