The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute git commands for change detection and mcp-cli to control browser sessions via the tabz MCP server. These operations are aligned with the intended purpose of automating visual QA smoke tests.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its interaction with untrusted external data.
Ingestion points: Browser console logs are ingested using the tabz_get_console_logs function in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard potential AI commands embedded within the logs.
Capability inventory: The skill has access to Bash, Read, Grep, and Glob tools, which could be leveraged if an injection is successful.
Sanitization: The skill lacks any visible mechanisms to sanitize or escape the content retrieved from the browser before it enters the agent's context.

visual-qa