agent-isles-render
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to use
npxto download and execute theagent-islespackage from the npm registry if it is not already present in the local project environment. - [COMMAND_EXECUTION]: The workflow involves executing shell commands through
npm,yarn, andpnpmto render Markdown files, watch for changes, and run diagnostic helpers. - [REMOTE_CODE_EXECUTION]: The use of
npx agent-isles@nextfacilitates the remote fetching and execution of code at runtime to perform its primary rendering function. - [DATA_EXPOSURE]: The skill reads local filesystem content (Markdown documents) and writes generated HTML artifacts back to the project directory.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external Markdown data which is subsequently read and verified by the agent. To mitigate risks, the instructions explicitly recommend the use of
--mode sanitizedwhen handling untrusted or mixed-trust documents.
Audit Metadata