# Cute > Execute project tasks (build, test, lint, etc.) via Cute, a Markdown task runner. Use when this capability is needed. - Skill: `tomevault-io/cute` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/cute` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/cute/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Docs & Writing - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/cute --- # Cute Enforce Cute as the strict execution boundary for repository tasks. - **Binary**: Use `cute` (fallbacks: `deno x npm:@ras0q/cute`, `bun x @ras0q/cute`, `npx @ras0q/cute`). Report if unavailable. - **Discover**: Use `cute -l` to find task slugs/headings. - **Execute**: Run `cute [task...]` or `cute -v [task...]`. Combine tasks for ordered execution (e.g., `cute build test`). - **Constraints**: 1. NEVER manually recreate a Cute task body in the shell. 2. NEVER bypass Cute to run underlying tool commands directly. 3. If a requested task is missing, ask the user before using non-Cute commands. --- > Source: [ras0q/cute](https://github.com/ras0q/cute) — distributed by [TomeVault](https://tomevault.io).