Gherkin story author. Collects a feature description from the human, writes a complete Gherkin story file to docs/stories/, and halts for human approval. The story IS the spec — no intermediate artifacts. Runs before DISCOVER.
Produces high-fidelity UI component mockups from approved wireframes and design tokens — React/HTML code for web targets, lipgloss-annotated terminal mockups for tui targets. Uses the mockup skill. Requires wireframe approval before starting.
Synthesises user research from problem statements. Produces personas, journey maps, and research summaries. Feeds the wireframe-architect and product-owner agents.
Produces brand palette, typography, spacing, and radius scales from a brief or keywords. Outputs palette.json, typography.json, and seeds tokens.json. Constrains to terminal-safe colors for tui targets. Uses the visual-identity skill. All outputs require human approval.
Produces low-fidelity wireframes from user stories and UX research. ASCII, SVG, HTML, or terminal output. Uses the wireframe skill. Every wireframe requires human approval before mockup proceeds.
Runs before the agent uses a tool for Bash, Write and Edit tool calls, executing pre-bash.sh and pre-write.sh via bash with --show-toplevel (3 commands). From kinncj/MAPLE.
Runs after a tool call finishes for Write, Edit and Bash tool calls, executing post-write.sh and post-bash.sh via bash with --show-toplevel (3 commands). From kinncj/MAPLE.
Extract Gherkin scenarios from story markdown files into runnable .feature files and generate step definition stubs. Use when syncing stories to test suites.
Read and write design tokens in W3C DTCG format (tokens.json) and emit CSS, Tailwind, and Mantine outputs. Use when modifying or generating design tokens.