setup-experiment

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). Most links point to official docs and GitHub, but the direct shell installer at https://cli.promptingco.com/install.sh is suspicious because it's a direct .sh download hosted on a different/shortened domain (potential typosquat or third‑party host) — a common malware distribution pattern.

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow includes “Step B1 — Capture experiment context” where it may perform web search and then “Fetch docs from the location captured in Step B1” to extract capability surface from product documentation/web pages, which is outsider-authored free text that the agent would read and include in its own LLM context.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly directs running remote code at runtime with "curl -fsSL https://cli.promptingco.com/install.sh | bash", which fetches and executes external installer content required for the CLI prerequisite.