.gitignore

# Miscellaneous
*.class
*.log
*.pyc
*.swp
.DS_Store
.atom/
.build/
.buildlog/
.history
.svn/
.swiftpm/
migrate_working_dir/

# IntelliJ related
*.iml
*.ipr
*.iws
.idea/

# The .vscode folder contains launch configuration and tasks you configure in
# VS Code which you may wish to be included in version control, so this line
# is commented out by default.
#.vscode/

# Flutter/Dart/Pub related
**/doc/api/
**/ios/Flutter/.last_build_id
.dart_tool/
.flutter-plugins-dependencies
.pub-cache/
.pub/
/build/
/coverage/

# Symbolication related
app.*.symbols

# Obfuscation related
app.*.map.json

# Android Studio will place build artifacts here
/android/app/debug
/android/app/profile
/android/app/release
.artisan/

# Local sibling-package path wiring (dev only)
pubspec_overrides.yaml

# Secrets (never commit; .env.example is the template)

# ac orchestration artifacts (plan + evidence)
.ac/

# Local QA report (not part of the published example)
/REPORT.md

# pubspec.lock IS committed here, reversing this project's earlier policy of
# ignoring it. That argument was correct about the failure mode (a lock generated
# with pubspec_overrides.yaml active encodes `source: path` entries carrying the
# ABSOLUTE sibling paths that file holds, which exist on exactly one machine)
# but drew the wrong conclusion from it: the fix
# for a lock generated the WRONG way is to generate it the right way, not to
# stop tracking it and hand every fork a first build that resolves whatever
# the caret ranges in pubspec.yaml happen to solve to on the day it is cloned
# rather than what CI actually proved green. `run_lockfile` in bin/check is
# what makes tracking safe: it fails the STAGED lock when it carries a
# `source: path` entry, so a lock generated with the overrides live never
# reaches main.

# Worktrees Claude Code creates for parallel sessions and isolated subagents.
# Only this subdirectory: the rest of `.claude/` is tracked, because the rules
# under it are part of the repository.
#
# They sit inside the working tree rather than beside it. `git clean -xdf` leaves them
# alone: git's nested-repository guard resolves a gitfile too, so it reports "Skipping
# repository .claude/worktrees/<slug>" and the contents survive. `git clean -xdff` does
# not skip them, and wipes a live worktree while `.git/worktrees/<slug>` survives,
# leaving a registration that `git worktree prune` then has to clear.
.claude/worktrees/