Prompt file imported from jlaws/dotfiles (
.codex/prompts/j-lang.md). Fill in{{arguments}}before use. Copyright stays with the author.
Before starting, gather diagnostic context:
- Detect primary language from config files (package.json, pyproject.toml, Cargo.toml, go.mod, Package.swift, Gemfile).
- Identify tooling by searching for linter configs, formatter configs, build tool settings.
- Check project conventions for existing patterns, module structure, and coding standards.
- Get scope overview of the target area (if {{arguments}} specifies a topic, scope to that; otherwise detect language from the project root).
For deep language guidance, delegate to the language-specialist agent, passing the detected language/tooling and the request. It loads its skills (language-testing-patterns, test-driven-development) and the .agents/references/languages/ library, then returns idiomatic guidance. Verify its output before presenting.
Help with: {{arguments}}