aspire-cli

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill instructs AI agents to call MCP tools that return environment variables/connection strings and even includes examples embedding plaintext passwords in CLI commands (e.g., playwright-cli fill "password123"), so an agent would likely receive and could be expected to include secret values verbatim in outputs.

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

• Third-party content exposure detected (high risk: 0.80). The skill's AI-testing workflow (SKILL.md "playwright-cli E2E testing" and the Aspire vs Playwright reference) explicitly instructs agents to use playwright-cli to open and scrape arbitrary URLs (it states Playwright-CLI can run against any accessible URL and "scrap[e] data from web pages") and the MCP tools (e.g., list_console_logs / list_structured_logs) return resource logs that the agent is expected to read and act on, meaning untrusted third‑party web/user-generated content can be fetched and can drive subsequent agent actions.