Frontend Playbook
Warn
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill's setup instructions direct the user or agent to install a plugin from an external GitHub repository (
github.com/austintgriffith/ethskills) and a package from an unverified registry (clawhub install ethskills). - [COMMAND_EXECUTION]: The
catalog.jsonmanifest specifies an installation command usingcurlto fetch content from an external domain (ethskills.com). - [PROMPT_INJECTION]: The skill instructs the agent to browse and follow instructions from an external URL (
https://ethskills.com/frontend-playbook/SKILL.md). This creates a surface for indirect prompt injection, as the content at that URL is not part of the analyzed package and could contain malicious instructions. - [REMOTE_CODE_EXECUTION]: While the fetched URL currently points to a markdown file, the inclusion of remote fetching in the
installandsetupfields introduces a mechanism where the source could be swapped for executable content.
Audit Metadata