webapp-testing
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The helper script
scripts/with_server.pyaccepts and executes arbitrary shell commands for starting servers and running automation tasks. It usessubprocess.Popenwithshell=Truefor the server commands andsubprocess.runfor the test command provided via command-line arguments. - [DYNAMIC_EXECUTION]: The instructions in
SKILL.mddirect the agent to write custom Python scripts using the Playwright library and execute them to automate interactions with web applications. - [INDIRECT_PROMPT_INJECTION]: The skill processes data from web applications (HTML content, selectors, and console logs), which constitutes an attack surface where a malicious web page could attempt to influence the agent's behavior.
- Ingestion points: Rendered DOM content accessed via
page.content(), element discovery via locators, and browser console logs captured inexamples/console_logging.py. - Boundary markers: None are present; the instructions do not include delimiters or specific warnings to ignore instructions found within the scraped content.
- Capability inventory: The agent can execute shell commands through the bundled
with_server.pyscript and perform file operations such as writing logs and screenshots. - Sanitization: There is no mention of sanitizing or validating the ingested web content before it is used to determine subsequent actions.
Audit Metadata