eliza
Pass
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches framework source code and CLI tools from the official elizaOS GitHub repository and the npm registry.\n- [COMMAND_EXECUTION]: Instructs users to use standard package management and build commands to initialize and run the agent runtime environment.\n- [PROMPT_INJECTION]: The framework documents an attack surface for indirect prompt injection where untrusted external data from connected platforms could influence agent behavior.\n
- Ingestion points: Processes user-provided text from Discord, Telegram, Twitter, and Farcaster clients.\n
- Boundary markers: The provided character and action examples do not illustrate the use of delimiters or specific isolation instructions for user input within the prompts.\n
- Capability inventory: Includes capabilities for high-impact actions such as Solana token transfers and swaps, as well as automated social media posting.\n
- Sanitization: The documentation snippets do not highlight explicit input sanitization or validation logic for the ingested messages.
Audit Metadata