eliza

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text from platform/user messages (e.g., Discord/Telegram/direct API message content) into the agent’s context for response generation.

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

• Potentially malicious external URL detected (high risk: 0.90). The troubleshooting doc explicitly recommends running "curl -fsSL https://bun.sh/install | bash", which fetches and executes remote code from https://bun.sh/install and bun is a required dependency for this project, so this is a high-confidence runtime/installation URL that can execute remote code.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The documentation explicitly includes a Solana integration/plugin that provides wallet management and actionable token operations. It lists concrete actions such as SEND_TOKEN (transfer SPL tokens), SWAP_TOKEN (token swap via aggregator), and STAKE_SOL, and shows storing a SOLANA_PRIVATE_KEY in secrets/env. These are specific crypto wallet/transaction capabilities (signing/sending transactions), not generic tooling — therefore the skill grants direct financial execution authority.