codexspec:constitution

Language Preference

Read .codexspec/config.yml. Two independent language controls apply (each falls back to language.output, then English):

• Interaction language (language.interaction): language for all conversation with the user — questions, explanations, status messages, and codexspec CLI terminal output.
• Document language (language.document): language for generated artifact files (requirements/spec/plan/tasks).

Converse in the interaction language and author artifacts in the document language. Apply the project's translation standard to both: translate by meaning (not word-for-word), keep English for terms with no good native equivalent, and write as if originally in that language.

User Input

the text after the $codexspec:constitution skill mention

Input Mode Detection

Parse the text after the $codexspec:constitution skill mention to determine execution mode:

Mode A: No Arguments → Present exploration options and let user choose

Mode B: Exploration Keyword (quick or deep) → Execute the corresponding exploration mode

Mode C: Project Principles Description → Use provided principles directly, skip exploration

---

Mode A: No Arguments - Present Options

IMPORTANT: Use the host agent's structured-question tool (e.g., AskUserQuestion or request_user_input) to present structured choices.

{
  "questions": [{
    "question": "I will explore the project to generate the constitution. Please select exploration depth:",
    "header": "Exploration",
    "options": [
      {
        "label": "quick",
        "description": "Config files + README + core entry points (~5-10 files), best for newly initialized projects"
      },
      {
        "label": "deep",
        "description": "Above + full source code analysis + architecture patterns, best for mature projects"
      },
      {
        "label": "Describe principles",
        "description": "Skip exploration, I will tell you the project principles directly"
      }
    ]
  }]
}