daytona-electron-den
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell scripts located in the
.devcontainer/directory and uses thedaytona execcommand to run processes within the sandbox environment. These actions are performed to orchestrate the server and client setup for testing purposes. - [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to capture the inner text of the Electron application (
document.body.innerText) to validate its state. This ingestion of untrusted application data creates a surface where embedded instructions could influence the agent's behavior. - Ingestion points: Captured UI text from the Electron application window via JavaScript execution.
- Boundary markers: No delimiters or specific instructions to ignore embedded content are provided in the prompts.
- Capability inventory: The agent has access to shell execution (
bash), network utilities (curl), and container management tools (daytona). - Sanitization: The skill does not implement any filtering or sanitization of the application content before it is processed by the agent.
Audit Metadata