v0.1.6

New: built-in docs — the docs and explain subcommands

import-lint docs [topic] prints a condensed built-in guide: concepts (the mental model), config (every option with its default), or fixing (how to fix a violation, in order of preference). import-lint explain <message-id> explains one diagnostic — what it means and the recommended fixes — for package, package:reexport, private, private:reexport, and unresolved (the ids reported in --format json's messageId field). Run either with no argument for an index.

The text ships inside the binary, so it always matches the installed version. It's written for humans and equally aimed at AI coding agents working in a repository that uses ImportLint.

New: a copy-pastable agent skill

skills/import-lint/ is a ready-made Agent Skill that teaches AI coding agents ImportLint's mental model, CLI usage, and the fix procedure for violations — including steering them away from tag-everything-@public "fixes". Copy it into your project (e.g. .claude/skills/import-lint/ for Claude Code, or wherever your agent tool looks for skills); it defers to the docs/explain subcommands above for version-matched depth.

import-lint init now ends with tips pointing at the built-in docs and the skill.

Docs

The Adoption guide was reworked, with smaller adjustments to the Concepts guide and Tutorial.

No lint-behavior changes — checking and output are identical to v0.1.5.

Full Changelog: v0.1.5...v0.1.6