eip
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to performgit cloneoperations on Ethereum specification and client repositories. It also utilizesgrepandgit logto search for EIP references within the downloaded codebases and commit history. - [EXTERNAL_DOWNLOADS]: Downloads source code from multiple public GitHub repositories, including official Ethereum specifications and major consensus and execution layer client implementations such as Geth, Prysm, and Lighthouse. These references target well-known open-source projects within the Ethereum ecosystem.
- [DATA_EXFILTRATION]: Performs network requests using
WebFetchandWebSearchto retrieve information fromeips.ethereum.org,forkcast.org, andethresear.ch. These operations are consistent with the skill's primary purpose of aggregating Ethereum-related technical data. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it ingests untrusted data from external web sources (EIP content and forum posts) and processes it using the agent's reasoning capabilities.
- Ingestion points: Data retrieved via
WebFetchfrom external websites and repositories. - Boundary markers: None identified; the instructions do not include specific delimiters or warnings to ignore instructions embedded in the external content.
- Capability inventory: The skill has access to powerful tools including
Bash,Task, and file system utilities. - Sanitization: No explicit sanitization or filtering of external content is performed before interpolation into the agent's context.
Audit Metadata