aspire

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md (section "1. Researching Aspire Documentation") and references/mcp-server.md explicitly instruct the assistant to call MCP tools that fetch and search public documentation on aspire.dev and GitHub (and fall back to Context7/other community sources), and the MCP tools also let the agent read logs/traces and execute resource commands—public/untrusted content the agent must interpret at runtime and that can materially change actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill contains installation commands that fetch-and-execute remote scripts (curl -sSL https://aspire.dev/install.sh | bash and irm https://aspire.dev/install.ps1 | iex) which install the Aspire CLI—a required dependency for MCP features—so these URLs cause remote code execution at setup/runtime.