Contract Addresses
Pass
Audited by Gen Agent Trust Hub on Jul 5, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill's installation process and setup instructions in
catalog.jsonutilizecurlto fetch content fromhttps://ethskills.com/addresses/SKILL.md. This introduces a dependency on an external, third-party domain that is not a well-known service or trusted organization. - [EXTERNAL_DOWNLOADS]: The
setupinstructions recommend installing a plugin from a personal GitHub repository (github.com/austintgriffith/ethskills), which constitutes an unverifiable dependency from a source not recognized as a trusted vendor. - [PROMPT_INJECTION]: The skill instructs the agent to read and follow the contents of an external URL (
https://ethskills.com/addresses/SKILL.md) before performing its primary task. This creates a surface for indirect prompt injection. - Ingestion points: External content fetched via
curlas defined incatalog.jsonand referenced inSKILL.md. - Boundary markers: Absent; the instructions do not provide delimiters or warnings to ignore embedded instructions in the fetched content.
- Capability inventory: The agent is instructed to read external content and use the provided contract addresses for blockchain operations, which is a high-stakes environment.
- Sanitization: Absent; there is no validation or filtering of the remote content before it is incorporated into the agent's context.
Audit Metadata