agent-tools

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). Most links are documentation pages on a single domain and not inherently malicious, but the published installation pattern (curl https://cli.inference.sh | sh) is a high‑risk distribution vector because it fetches and executes a remote shell script without inspection, so the domain could be used to deliver malware if compromised.

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 explicitly shows running web-search apps (e.g., "infsh app run tavily/search-assistant --input '{"query":"latest AI news"}'") and lists "Web search (Tavily, Exa)" which indicates the agent will ingest public web/social content as part of its workflow, allowing untrusted third-party content to influence actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs installing a required CLI at runtime using a shell pipe that fetches and executes remote code ("curl -fsSL https://cli.inference.sh | sh"), so the URL https://cli.inference.sh is a runtime external dependency that executes remote code.