What problem does it solve? Verifying interactive terminal applications, CLI wizards, and TUI flows is hard to automate because prompts, inputs, outputs, and exit codes must be captured and judged against acceptance criteria. This Skill drives terminal applications like a user and produces transcripts, exit-code evidence, and pass/fail evaluations inside the Agentic SDLC workflow. ## Core Features & Use Cases - Pexpect-Style Interaction: Launch a CLI or TUI, wait for expected prompts, send inputs, and capture output and exit codes. - Evidence Capture: Record terminal transcripts, exit-code summaries, output assertions, and side-effect validation as run evidence. - Failure Classification: Map prompt mismatches, harness timeouts, and incorrect expected transcripts to evaluation, environment, or test defect categories. - Use Case: During a UAT evaluation of a CLI installer wizard, the coordinator routes to this Skill to drive the wizard, capture the full transcript, verify generated files, and return a pass/fail result against the acceptance criteria. ## Quick Start Ask the Agentic SDLC coordinator to evaluate the CLI wizard against its acceptance criteria using sdlc-tui-test and record the transcript and exit code as evidence.
