addresses
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and directs the agent to fetch data from external GitHub repositories:
https://github.com/monad-crypto/protocolsandhttps://github.com/monad-crypto/token-list. These are used as supplementary data sources for smart contract addresses and token information. - [COMMAND_EXECUTION]: The skill provides example shell commands using the
castutility (from the Foundry toolkit) to verify contract bytecode. These commands (cast code [address] --rpc-url [url]) are intended for verification and auditing purposes. - [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and process data from external JSON and JSONC files within the
monad-cryptorepositories. This creates a surface for indirect prompt injection if those external data files were to contain malicious instructions instead of the expected smart contract address data. - Ingestion points: Files like
protocols/mainnet/clober.jsoncandprotocols/testnet/[protocol_name].jsonfrom external GitHub repositories. - Boundary markers: None explicitly defined; the skill relies on the structured nature (JSON) of the external files.
- Capability inventory: The agent is expected to read these files to retrieve contract addresses for user interaction.
- Sanitization: No explicit sanitization of the values fetched from the external JSON files is described.
Audit Metadata