QA Checklist
Warn
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill is configured to fetch its primary instructions and content dynamically from
https://ethskills.com/qa/SKILL.mdusing acurlcommand at install time. - [REMOTE_CODE_EXECUTION]: The
catalog.jsonfile provides setup instructions that direct users to install external plugins usingclaude plugin install https://github.com/austintgriffith/ethskillsandclawhub install ethskills. These actions download and integrate third-party code into the agent's environment. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to audit Ethereum dApps, which involves ingesting external application code into the agent's context. The instructions do not define sanitization logic or boundary markers (e.g., XML tags or delimiters) to differentiate between the audit guidelines and potentially malicious instructions embedded in the audited code.
- Ingestion points: Audited dApp code files read by the agent.
- Boundary markers: None detected in the skill instructions.
- Capability inventory: Reading files and providing audit feedback.
- Sanitization: None detected; the skill relies on the agent to interpret the ingested code as data only.
Audit Metadata