awn

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). These URLs are high-risk because they instruct executing a remote install.sh via curl|bash from a raw GitHub URL (remote shell execution) and point to an unverified aggregator (claudskills.com) plus a custom gateway—installing a daemon that creates persistent keys and opens network ports from unreviewed/unknown sources can easily distribute malware or backdoors unless you manually review and verify the upstream repository and binaries.

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow includes awn worlds / awn join, which queries the external Gateway at https://gateway.agentworlds.ai and resolves/stores world co-member endpoints; any returned world/server metadata is outsider-authored free text that the daemon/CLI would ingest at runtime to proceed.

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

• Potentially malicious external URL detected (high risk: 0.90). The install step uses curl -fsSL https://raw.githubusercontent.com/ReScienceLab/agent-world-network/main/packages/awn-cli/install.sh | bash, which fetches and executes a remote shell script as a required installation step (remote code execution).