The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is configured to execute npx playwright commands for running and debugging E2E tests. This capability is restricted to specific npm and npx scopes within the configuration, aligning with the skill's primary functional purpose.
[PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection because its core workflow involves reading external documentation to generate code.
Ingestion points: The skill reads UX/UI specifications and user flow descriptions using the Read and Glob tools.
Boundary markers: The provided templates use standard Markdown formatting; however, they lack explicit security delimiters or instructions to ignore potential commands embedded within ingested specifications.
Capability inventory: The skill has the ability to create and modify files (Write, Edit) and execute shell commands (Bash), which could be abused if malicious instructions are present in the ingested data.
Sanitization: No specific validation, filtering, or escaping logic is defined in the instructions for handling the content of external specifications.
[SAFE]: No obfuscation techniques, unauthorized network exfiltration, or persistence mechanisms were detected. The credentials used in code examples (test@example.com, password123) are clearly marked as placeholders and do not represent a security risk.

qa-interaction