Conversation
- AGENTS.md: shared guide for AI coding agents (commands, layout, conventions, testing patterns, boundaries, open work). CLAUDE.md imports it. - docs/: architecture (data flow, result object, exit codes), verified known issues/tech debt, and an adding-a-validator recipe. - .claude/: add-validator skill and shared permission settings. - README: supported table now lists engine keys and commands (adds missing adb/windows, drops stale semver column); fix programmatic example and result-object description (status stays 0 on invalid environments). - lib/readme.spec.js: fail if a validator is missing from the README table. - package.json: add "files" so only bin/ and lib/ runtime files are published. - .gitignore: ignore local-only Claude Code files. Co-Authored-By: Claude Opus 5.5 <[email protected]>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Sets the repo up so AI coding agents (Claude Code, Copilot, Codex, Cursor) can work on it safely. No changes to runtime code.
Agent guidance
AGENTS.md: the main guide for every agent. Covers commands, layout, code style, testing patterns, rules for agents, and open issues/PRs. Treats CLI output, exit codes and the result object as public API. Agents must never publish or version, and never bumpcolorspast 1.4.0.CLAUDE.md: importsAGENTS.mdplus a few Claude-only notes..claude/skills/add-validator: the steps for adding a new tool..claude/settings.jsonallows test, lint and read-onlyghcommands, and blocksnpm publish/npm version.Knowledge docs
docs/architecture.md: how data flows, the result object, exit codes, and how the test stubs work.docs/known-issues.md: quirks confirmed by running the CLI, each marked as safe to fix or a breaking change. Examples:statusis0even when the environment is invalid; engine keys with no validator fail the whole check; results come back in nondeterministic order; and a Bluebird →util.promisifyswap would pass the tests but break real runs.docs/adding-a-validator.md: recipe for adding a validator.README fixes
engineskey and the command run. It adds the missingadbandwindowsrows and drops the stale semver column; all validators already support semver ranges.status !== 0, which never catches an invalid environment. Both are fixed, and the result-object description is corrected.npm ciinstead of installing ESLint globally.Guardrails
lib/readme.spec.js: fails if a validator key has no row in the README table.package.jsonfiles: the published package now contains onlybin/andlib/runtime files (plus README/LICENSE/package.json).npm pack --dry-run..gitignore: ignores.claude/worktrees/and.claude/settings.local.json.Test plan
npm test: 118 assertions pass (Node 24)npm run lint: cleanreadme.spec.jsfails with a clear message and a non-zero exitnpm pack --dry-runlists only runtime filesRelated: builds on the context in draft #108 and #109 without overlapping their changes.
🤖 Generated with Claude Code