The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes various shell commands for version control (git checkout, git commit) and project-specific testing, linting, and formatting tools. These operations are core to its TDD functionality but involve direct interaction with the host environment.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and acts upon instructions from external plan files or user input without proper isolation. * Ingestion points: Plan files are read via the Read tool, and user input is gathered via AskUserQuestion for behavior identification. * Boundary markers: No delimiters or instructions are used to isolate or ignore malicious instructions that might be embedded within the ingested plan data. * Capability inventory: The agent has the ability to write code to files, manage git repositories, and execute arbitrary project-defined test suites and linters. * Sanitization: No validation, filtering, or sanitization of external data is performed before it is used to drive the autonomous TDD cycles.

tdd-execute