nano-block-lattice-expert

Pass

Audited by Gen Agent Trust Hub on May 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch the xno-skills package and the @open-wallet-standard/core library from public npm registries during execution.
  • [REMOTE_CODE_EXECUTION]: The instructions mandate the use of the @latest tag (e.g., bunx -y xno-skills@latest) when executing commands. This practice runs unpinned code from a remote source, creating a supply chain vulnerability where an update to the package could change the agent's behavior without review.
  • [COMMAND_EXECUTION]: The skill provides explicit patterns for the agent to execute shell commands using bunx, pnpm, or npx to interact with the protocol and manage wallets.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes data from the Nano blockchain, such as transaction history and account states via RPC calls.
  • Ingestion points: Account history and state verification via RPC (history, account-balance).
  • Boundary markers: Absent. The instructions do not specify the use of delimiters or warnings to ignore instructions embedded in transaction data.
  • Capability inventory: The agent has access to shell execution (xno-skills CLI) and wallet operations (ows).
  • Sanitization: None mentioned for data retrieved from the blockchain ledger.
Audit Metadata
Risk Level
SAFE
Analyzed
May 3, 2026, 05:24 AM