dev-swarm-install-ai-code-agent

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs performing web searches for "other AI code agents" (SKILL.md Step 1) and its reference files include curl commands that fetch and pipe remote install scripts (e.g., references/claude-code.md and references/copilot-cli.md), meaning the agent will retrieve and act on open/public third-party content that could carry injected instructions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill includes installation commands that are run at runtime which fetch and execute remote scripts (curl | bash or irm | iex), specifically https://claude.ai/install.sh, https://claude.ai/install.ps1, https://claude.ai/install.cmd, and https://gh.io/copilot-install, so the fetched content will execute remote code and is presented as a required installation option.