migrator-uniswap-v3
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides explicit bash commands in
references/INTERFACE.mdto extract smart contract ABIs using theforgeCLI tool andjq. While these are legitimate developer tools, they represent a vector for executing commands based on project documentation. - [EXTERNAL_DOWNLOADS]: The skill points to external resources on GitHub (
github.com/whetstoneresearch/doppler) as the primary source of truth for contract addresses and source code inSKILL.md,references/INTERFACE.md, andreferences/OVERVIEW.md. - [PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection by instructing the agent to fetch and process external source code from GitHub to perform verification tasks.
- Ingestion points: External Solidity files and deployment JSONs referenced via GitHub URLs in
references/INTERFACE.mdandreferences/OVERVIEW.md. - Boundary markers: No explicit markers are used to isolate or warn the agent about potentially malicious instructions embedded in the external Solidity comments or metadata.
- Capability inventory: The skill leverages
forgefor contract inspection and describes logic for verifying on-chain transaction outcomes. - Sanitization: There is no evidence of sanitization or content validation for the fetched external source files.
Audit Metadata