product-hunt-launch

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). These URLs point to an installer script (cli.inference.sh) and a binary distribution/checksums host (dist.inference.sh) for a relatively unknown domain — the presence of checksums is a mitigant, but piping a remote script to sh and downloading platform binaries from a non-established domain is inherently risky and could be used to distribute malware if the site or artifacts are compromised.

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

• Third-party content exposure detected (high risk: 0.90). SKILL.md explicitly instructs the agent to run infsh apps like "tavily/search-assistant" and "exa/search" (see Quick Start and Research for Preparation) to fetch and analyze public Product Hunt/search results and community sentiment, i.e., untrusted third‑party web/user-generated content that the agent is expected to read and that can materially influence launch decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's Quick Start explicitly runs a remote install command that fetches and executes code via "curl -fsSL https://cli.inference.sh | sh" (and the infsh CLI then invokes remote apps hosted under inference.sh/dist), so https://cli.inference.sh is a runtime external dependency that executes remote code.