contract-decode
Pass
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes contract ABIs and transaction data fetched from external APIs and user input. This ingestion of untrusted data creates a surface for indirect prompt injection, particularly as the data is used to populate arguments for shell-based decoding tools.\n
- Ingestion points: Contract ABIs fetched from Sourcify and Etherscan; hexadecimal strings provided by the user in messages.\n
- Boundary markers: The instructions do not include specific delimiters or 'ignore' directives to protect the agent from instructions that might be embedded in fetched contract metadata or names.\n
- Capability inventory: The skill utilizes the
Bashtool to runcast,curl, andpython3, and usesWebFetchfor network operations.\n - Sanitization: The workflow relies on JSON parsing and command timeouts, but does not explicitly sanitize text strings extracted from external ABIs before they are passed into shell loops.\n- [EXTERNAL_DOWNLOADS]: Fetches data from well-known blockchain services such as Etherscan, Sourcify, and 4byte.directory to retrieve contract ABIs and signature candidates.\n- [COMMAND_EXECUTION]: Executes local commands using the
castCLI andpython3to perform ABI decoding and cryptographic signature generation.
Audit Metadata