radius-dev

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process data from the Radius blockchain and transaction cost APIs.
  • Ingestion points: The skill fetches network transaction costs and reads contract state/logs via viem and wagmi calls to radiustech.xyz RPC endpoints.
  • Boundary markers: The provided code examples do not include explicit instructions or delimiters to ignore potential malicious content embedded in blockchain data.
  • Capability inventory: The skill utilizes network access for blockchain interactions and facilitates the execution of smart contract transactions.
  • Sanitization: There is no evidence of specific sanitization or filtering logic applied to data retrieved from the blockchain before it is used in the agent's context.
  • [EXTERNAL_DOWNLOADS]: The skill connects to external network resources to perform its primary functions.
  • Fetches transaction cost data from https://testnet.radiustech.xyz/api/v1/network/transaction-cost and https://network.radiustech.xyz/api/v1/network/transaction-cost.
  • Interacts with blockchain RPC endpoints at https://rpc.testnet.radiustech.xyz and https://rpc.radiustech.xyz.
  • These downloads/references target the official infrastructure of the Radius project and are considered neutral findings under the intended purpose of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 01:15 PM