This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
A curated, progression-ordered learning path for Canton Network developers, published at https://github.com/BootNodeDev/canton-learning-path.
Content lives in README.md (title, intro, Foundations, chapter index, License) and per-chapter files under docs/ (concepts.md, daml.md, building.md, internals.md, operations.md, ai-assisted.md, community.md). Everything else (LICENSE, .gitignore, .claude/) exists to support those. There is no build system, no test suite, no package manager, no application code.
All editorial rules, section order, formatting requirements, lint checks, and workflow live in .claude/skills/maintain-canton-learning-path/SKILL.md. When adding, removing, or updating entries, read that skill first and follow it. This file intentionally does not duplicate those rules so they stay in one place.
Invoke the skill via the Skill tool when the task is: add/remove/update a resource, refresh links, or reorganize a section.
Every URL in README.md and docs/*.md must resolve and contain content matching its description. Use WebFetch to verify before adding. The project .claude/settings.json pre-authorizes WebFetch against the Canton-ecosystem domains already in use; if you need a new domain, add it to that file (tracked) rather than to .claude/settings.local.json (ignored, personal overrides only).
The skill's SKILL.md contains the exact grep/awk commands for checking forbidden parenthetical attributions, em/en dashes, emojis, and the 10-link-per-section cap. Run them before every commit.
Single main branch, linear history. Changes ship as small, focused commits pushed directly to main. No PR workflow is enforced on this repo.