The Agent Skills Directory

[PROMPT_INJECTION] (HIGH): The skill contains a 'black-box' instruction: 'DO NOT read the source until you try running the script first... They exist to be called directly as black-box scripts rather than ingested into your context window.' This is a direct attempt to bypass security review and AI oversight of executable code.
[COMMAND_EXECUTION] (HIGH): The scripts/with_server.py utility is designed to execute arbitrary shell commands provided via the --server flag (e.g., npm run dev, python server.py). This creates a significant risk of command injection if the server strings are influenced by untrusted data.
[REMOTE_CODE_EXECUTION] (HIGH): The skill facilitates the creation and execution of Playwright automation scripts. Because these scripts interact with web content, they are vulnerable to indirect prompt injection where a malicious web page could trigger the agent to execute unauthorized logic within the Playwright context.
[DATA_EXFILTRATION] (MEDIUM): The skill provides patterns for capturing DOM content and screenshots (e.g., /tmp/inspect.png). In the absence of strict output controls, this capability can be used to capture and exfiltrate sensitive data from local-network applications or user sessions.

webapp-testing