* fix(shell-completion): resolve ENOTDIR error when parent path is a file
Fixes issue where `--shell-completion` fails with ENOTDIR error when
a file exists where a directory should be created (e.g., ~/.zsh exists
as a file instead of a directory).
Changes:
- Added ensureDirectory() helper that safely creates directories
- Validates parent paths are directories, not files
- Provides clear error message when file conflicts occur
- Applied to all shell installers (bash, zsh, fish, powershell)
- Maintains idempotent behavior (safe to call multiple times)
Before: mkdir with recursive:true fails silently with ENOTDIR
After: Clear error message guides user to resolve file conflict
* chore: bump version to 4.1.4
* feat(completion): improve UI/UX with descriptions and grouping
Improves shell completion UI/UX across all shells (bash, zsh, fish,
PowerShell) with better organization and clearer descriptions.
Changes:
- Zsh: Added descriptions for all profiles and grouped by category
- Commands: "auth", "doctor" with descriptions
- Model profiles: glm, glmt, kimi, etc. with descriptions
- Account profiles: Dynamic with "Account-based profile" label
- Used _alternative for visual grouping
- Fish: Added explicit completions with descriptions for known profiles
- Replaced generic profile listing with specific entries
- Each profile now shows clear description (e.g., "GLM-4.6 (cost-optimized)")
- Bash: Added --shell-completion flag and subflags completion
- PowerShell: Added --shell-completion flag and subflags completion
- All shells: Added completion for --shell-completion subflags
Before (zsh):
ccs
auth -- Manage multiple Claude accounts
doctor -- Run health check and diagnostics
default glm glmt kimi max
After (zsh):
commands
auth -- Manage multiple Claude accounts
doctor -- Run health check and diagnostics
model profiles
default -- Default Claude Sonnet 4.5
glm -- GLM-4.6 (cost-optimized)
glmt -- GLM-4.6 with thinking mode
kimi -- Kimi for Coding (long-context)
max -- Claude Opus (maximum capability)
account profiles
work -- Account-based profile
personal -- Account-based profile
Consistent, organized, and informative completion across all shells.
* fix(completion): handle custom profiles and fix zsh syntax errors
Fixes two issues:
1. Zsh syntax errors with _describe -t flag in _alternative
2. Adds general handling for custom settings profiles (e.g., m2)
Changes:
- Zsh: Fixed _alternative syntax (removed -t tag from _describe calls)
- Error was: "_describe:21: bad option: -M"
- Cause: Tag is auto-derived in _alternative, don't specify with -t
- Fish: Added __fish_ccs_get_custom_settings_profiles function
- Dynamically loads non-hardcoded profiles from config.json
- Shows "Settings-based profile" description for custom profiles
- Removed 'max' from hardcoded known profiles
- 'max' is user's account-based profile, not a default setting
Now supports any custom settings profile (e.g., m2.settings.json for
Minimax M2) without hardcoding. Custom profiles show with generic
"Settings-based profile" description.
Installation properly overwrites:
- fs.copyFileSync overwrites completion files by default
- RC files only modified if marker not already present
* fix(zsh): simplify completion to avoid _alternative syntax issues
Replaced _alternative with multiple _describe calls to fix persistent
zsh completion errors.
The _alternative approach was causing:
- '_describe:21: bad option: -M'
- '(eval):1: bad substitution'
- Unwanted variable expansion in completion menu
New approach uses sequential _describe calls which zsh handles correctly.
Simpler, more reliable, and still shows grouped completion with
descriptions.
Before: _alternative with complex nested _describe calls (broken)
After: Three simple _describe calls (works correctly)
* feat(zsh): add colors and improved formatting to completion
Enhances zsh completion UI/UX with colors and better spacing:
Colors:
- Blue: commands (auth, doctor)
- Green: model profiles (default, glm, glmt, kimi, custom)
- Yellow: account profiles
- Gray: descriptions
Formatting:
- Wider separator ( -- ) for better readability
- Group headers in bold cyan
- Menu selection enabled for navigation
- list-rows-first for better column layout
Before:
auth -- Manage multiple Claude accounts
default -- Default Claude Sonnet 4.5
doctor -- Run health check and diagnostics
glm -- GLM-4.6 (cost-optimized)
After:
commands (cyan header)
auth -- Manage multiple Claude accounts (blue)
doctor -- Run health check and diagnostics (blue)
model profiles (cyan header)
default -- Default Claude Sonnet 4.5 (green)
glm -- GLM-4.6 (cost-optimized) (green)
glmt -- GLM-4.6 with thinking mode (green)
kimi -- Kimi for Coding (long-context) (green)
account profiles (cyan header)
max -- Account-based profile (yellow)
Table-like appearance with colors matching --help output style.
* feat(fish): add colors to completion descriptions
Adds color-coded descriptions to Fish completion matching zsh style:
Colors:
- Blue: commands (auth, doctor)
- Green: model profiles (default, glm, glmt, kimi, custom)
- Yellow: account profiles
Fish has excellent native color support via set_color, making this
straightforward to implement.
Before:
auth Manage multiple Claude accounts
default Default Claude Sonnet 4.5
glm GLM-4.6 (cost-optimized)
max Account profile
After:
auth Manage multiple Claude accounts (blue)
default Default Claude Sonnet 4.5 (green)
glm GLM-4.6 (cost-optimized) (green)
max Account profile (yellow)
Consistent color scheme across zsh and fish shells.
* fix(zsh): correct color application for commands vs descriptions
Fixed color codes being applied backwards - commands were gray while
descriptions were colorful.
Issue:
- Used '1;34' format which zsh misinterpreted
- Commands appeared gray (unreadable)
- Descriptions appeared in color
Fix:
- Simplified color codes from '1;34' to '34' (just the color, no style prefix)
- Pattern now: =(#b)(command)(-- description)=34=90
- First group (command): color 34 (blue/green/yellow)
- Second group (description): color 90 (gray)
Before:
auth -- Manage multiple Claude accounts (gray) (blue)
After:
auth -- Manage multiple Claude accounts (blue) (gray)
Commands now properly colored, descriptions muted.
* feat(zsh): improve section header visibility with separators
Makes section headers more distinct and recognizable as dividers:
Before:
commands
auth -- ...
After:
── commands ──
auth -- ...
Changes:
- Added blank line before headers ($'\n')
- Bold yellow with decorative separators (── text ──)
- Clearly distinguishes section headers from actual commands
Now it's obvious that 'commands', 'model profiles', and 'account
profiles' are just section labels, not completable items.
* fix(zsh): use full ANSI color codes to fix reversed colors
The issue was using shorthand color codes (34, 90) instead of full
ANSI format (0;34, 2;37). Zsh was misinterpreting the shorthand,
causing colors to be applied to wrong groups.
Fixed:
- Commands: 0;34 (normal blue) - was appearing gray
- Model profiles: 0;32 (normal green) - was appearing gray
- Account profiles: 0;33 (normal yellow) - was appearing gray
- Descriptions: 2;37 (dim white) - was appearing colorful
Color codes must be escaped as 0\;34 in zsh strings.
Before fix:
auth (gray) -- description (blue)
After fix:
auth (blue) -- description (dim gray)
* fix(zsh): swap color order for completion groups
Zsh may apply colors in reverse order to capture groups on some systems.
Swapped color order: group 2 first, then group 1.
Pattern: (command)(separator + description)
Was: =blue=dim_white (applied as: blue to cmd, dim to desc)
Now: =dim_white=blue (applied as: dim to desc, blue to cmd)
Testing if this resolves the reversed color issue.
* fix(zsh): add empty leading color to fix reversed coloring
ROOT CAUSE: Zsh list-colors apply first color to WHOLE match, then
override with subsequent colors for each group. Without an empty
leading color, the whole match color leaks to groups without overrides.
Pattern behavior:
=(#b)(cmd)(desc)=BLUE=DIM
→ Whole: BLUE, Group1: DIM (override), Group2: BLUE (no override)
→ Result: cmd=dim, desc=BLUE (REVERSED!)
=(#b)(cmd)(desc)==BLUE=DIM
→ Whole: none, Group1: BLUE, Group2: DIM
→ Result: cmd=BLUE, desc=dim (CORRECT!)
The '==' at start means 'no whole-match color', preventing color bleed.
Now commands will be colorful (blue/green/yellow) and descriptions dim.
* refactor(help): remove specific account examples and generalize description
Removed deterministic account examples (work, personal, team) to make
the help text more generic and less prescriptive.
Changes:
- Removed 'ccs work' and 'ccs personal' example lines
- Removed 'Multi-account workflow' examples section
- Updated description from 'multiple Claude accounts (work, personal, team)'
to 'multiple Claude accounts and alternative models'
- Changed to 'Run different Claude CLI sessions concurrently'
- Applied consistently across Node.js (bin/ccs.js), bash (lib/ccs),
and PowerShell (lib/ccs.ps1)
This makes the help text more flexible and doesn't imply specific
use cases or account naming conventions.
* refactor(help): clarify delegation section and remove non-existent command
Updated delegation section in help text across all implementations:
Changes:
- Renamed section from 'Delegation (Token Optimization)' to
'Delegation (inside Claude Code CLI)' to clarify context
- Removed non-existent '/ccs:create m2' command
- Simplified description to focus on cost savings
- Updated command descriptions:
- '/ccs:glm' now 'for simple tasks' (clearer use case)
- '/ccs:kimi' unchanged (already clear)
- Added delegation section to PowerShell help (was missing)
- Consistent messaging across Node.js, bash, and PowerShell
The new section makes it immediately clear that delegation commands
are used within Claude Code CLI sessions, not as standalone commands.
---------
The 'ccs doctor' command was incorrectly checking for delegation commands
in ~/.ccs/shared/commands/ccs/ instead of ~/.ccs/.claude/commands/ccs/,
causing false "not installed" warnings even after successful installation.
Changes:
- Fix delegation check to look in correct directory (~/.ccs/.claude/)
- Remove check for non-existent create.md file
- Bump version to 4.1.3
- Remove 5 deprecated model fields from Kimi settings causing auth failures
- Update config/base-kimi.settings.json template with clean configuration
- Add automatic migration in scripts/postinstall.js to remove deprecated fields
- Preserve user API keys and custom settings during migration
- Update CHANGELOG.md with v4.1.1 and v4.1.2 entries
- Bump version to 4.1.2 across all files and installers
Fixes authentication issues with Moonshot AI (Kimi) API due to deprecated
model parameters that are now rejected by the service.
Fixes issue where npm install -g or ccs update would fail to copy
.claude/ directory to ~/.ccs/.claude/, causing symlink installation
errors. ClaudeDirInstaller now ensures source exists before
ClaudeSymlinkManager attempts linking.
Add defensive checks in result-formatter, headless-executor, and session-manager
to prevent TypeError when delegated sessions timeout without emitting a result.
Includes 4 comprehensive unit tests for undefined/null totalCost scenarios.
Fixes formatting error that prevented timeout messages from displaying.
- Reorganized README structure with collapsible sections for better readability
- Added comprehensive GLMT experimental warnings and troubleshooting guides
- Updated installation instructions with traditional vs npm options
- Added parallel workflow examples and detailed feature comparisons
- Enhanced documentation with proper acknowledgments and security limits
- Standardized language and formatting across all README variants
- Move GLMT section to proper position in English README
- Add missing npm badge to Vietnamese README
- Add GLMT and Architecture sections to all translations
- Update configuration examples with glmt profile
- Add Windows Developer Mode support documentation
- Fix documentation links across all language versions
- Add prominent NOT PRODUCTION READY warning for GLMT
- Update GLM vs GLMT comparison to reflect experimental nature
- Tone down overconfident marketing language in tool support
- Add honest failure rates for streaming functionality
- Reference CCR hustle with Transformer of Bedolla as alternative
- Update thinking keywords to reflect inconsistent behavior
Condensed comprehensive documentation into essential development guidance with:
- Simplified project overview and core principles
- Consolidated technical implementation details
- Streamlined GLMT debugging and troubleshooting section
- Focused development workflows and checklists
- Removed redundant sections and verbose explanations
Maintained all critical constraints and code standards while improving readability for developers.
Remove references to deprecated CCS_GLMT_FORCE_ENGLISH, CCS_GLMT_THINKING_BUDGET, and CCS_GLMT_STREAMING environment variables that were removed in v3.4.3.
Changes:
- README.md: Update GLMT environment variable documentation to reflect current intelligent control system
- tests/unit/glmt/locale-enforcer.test.js: Remove outdated test scenario for removed environment variable
- docs/glmt-controls.md: Rewrite documentation to describe new automatic control mechanisms
The GLMT component now uses intelligent, automatic controls instead of manual environment variable configuration.
- update require path from '../bin/shared-manager' to '../bin/management/shared-manager'
- fixes symlink creation failure in postinstall
- bump version to 3.4.4
- Add GLMT proxy server for GLM model routing
- Add GLMT transformer for output format conversion
- Update CLI with new proxy and transformer commands
- Add comprehensive test suite for new functionality
- Update documentation and architecture guides
- Bump version and update changelog
Fixed:
- Migration now runs during installation across all methods (npm, bash, PowerShell)
- Guarantees ~/.ccs/shared/ populated immediately with ~/.claude/ content
- Users no longer need to run ccs command to trigger migration
Changed:
- Refactored SharedManager with _needsMigration() and _performMigration() methods
- _copyDirectory() returns {copied, skipped} stats and preserves existing files
- Shows detailed migration output: '[OK] Migrated 5 commands, 19 skills'
- Removed lazy migration from bin/ccs.js, lib/ccs, lib/ccs.ps1
Implementation:
- npm: Migration in scripts/postinstall.js
- bash: Migration in installers/install.sh (migrate_shared_data function)
- PowerShell: Migration in installers/install.ps1 (Invoke-SharedDataMigration)
- Fixed arithmetic expansion with set -e (changed ((var++)) to var=$((var + 1)))
Cross-platform parity maintained across all installation methods.
Phase 1: Multi-profile shared data via symlinks
Added:
- SharedManager class for symlink orchestration (bin/shared-manager.js)
- Auto-migration from ~/.claude/ to ~/.ccs/shared/ on first run
- Shared directories: commands/, skills/, agents/
- Windows fallback: copies dirs if symlinks fail
Fixed:
- Migration logic now detects empty directories
- Previously skipped migration when postinstall created empty dirs
- Now properly copies from ~/.claude/ when shared dirs are empty
Changed:
- Instance initialization symlinks to shared dirs instead of copying
- Postinstall creates ~/.ccs/shared/ structure automatically
- All implementations (Node.js, bash, PowerShell) updated for consistency
- Help text includes agents/ in shared data section
Technical:
- Profile-specific data remains isolated (settings, sessions, todolists, logs)
- Migration is idempotent: safe to run multiple times
- Cross-platform symlink support with graceful fallback
Closes#4
This release fixes default profile behavior and streamlines help output.
Breaking Changes:
- Profile creation NO LONGER auto-sets as default
- Users must explicitly run `ccs auth default <profile>` to set default
- Without explicit default, `ccs` uses implicit default from ~/.claude/
Auth Default Behavior:
- Removed auto-set default logic in profile-registry.js
- Removed auto-set default in bash register_profile() function
- Removed auto-set default in PowerShell Register-Profile function
- Implicit 'default' profile always exists (uses ~/.claude/)
- Enhanced success messages guide users to set explicit default
- Updated auth help with examples and note about default behavior
Help Text Simplification:
- Removed lengthy Examples section from main help (~40% shorter)
- Condensed Account Management section to `ccs auth --help`
- Kept detailed examples in `ccs auth --help` where relevant
- Consistent across npm, bash, and PowerShell implementations
Files Changed:
- bin/profile-registry.js: Removed auto-default logic
- bin/auth-commands.js: Updated help and success messages
- bin/ccs.js: Simplified main help text
- lib/ccs: Fixed bash implementation + simplified help
- lib/ccs.ps1: Fixed PowerShell implementation + simplified help
- VERSION, package.json, installers/*: Version bump to 3.0.2
Fixes #TBD
- Add auto-recovery mechanisms for missing/corrupted configuration files
- Implement comprehensive health check command (`ccs doctor`)
- Enhance error messages with context-aware diagnostics and recovery commands
- Fix silent postinstall failures - now exits with proper error codes
- Add RecoveryManager class for automatic config restoration
- Add ErrorManager class for structured, helpful error messages
- Update postinstall script to validate created files and auto-create ~/.claude/settings.json
- Add doctor command support to bash and npm implementations
- Implement atomic file operations to prevent corruption
- Add comprehensive testing scenarios and validation
Fixes critical issue where npm install succeeded but CCS failed on first run.
Enhances user experience with automatic recovery and clear error guidance.
BREAKING CHANGE: Postinstall now exits with error code 1 on critical failures
* feat: implement native multi-account switching with isolated instances
Add account-based profile management system that enables users to run
multiple Claude accounts concurrently with complete isolation.
Key features:
- Profile registry (~/.ccs/profiles.json) tracks account profiles
- Instance isolation (~/.ccs/instances/<profile>/) for each account
- Auth commands: create, list, show, remove, default
- Backward compatible with settings-based profiles (GLM, Kimi)
- Auto-copies global .claude configs to new instances
Implementation:
- Phase 1: Profile detection logic (account vs settings-based)
- Phase 2: Instance management (initialization, validation)
- Phase 3: Auth CLI commands
- Phase 4: Profile registry CRUD operations
- Phase 5: Execution routing based on profile type
Both lib/ccs (bash) and lib/ccs.ps1 (PowerShell) updated for
cross-platform support.
* fix(ci): add pull_request trigger to satisfy branch protection
The branch protection rule requires "Deploy ccs-installer to CloudFlare"
status check to pass, but the workflow only ran on push to main branch.
This caused PRs to wait indefinitely for a status that never reported.
Changes:
- Add pull_request trigger with same path filters
- Split deployment into conditional steps:
- PR mode: dry-run validation only (--dry-run flag)
- Production: actual deployment (push to main only)
- Keeps same job name to satisfy branch protection requirement
Security:
- No deployment from PR branches (dry-run only)
- Production deploy only when push to main AND ref check
- Secrets used safely in both contexts
This fixes PR #3 which was stuck waiting for status.
* fix(ci): remove path filter from pull_request trigger
The path filter prevented workflow from running on PRs that don't
modify worker/installer files, causing required status check to
never report. This blocked PR #3 indefinitely.
Changes:
- Remove paths filter from pull_request trigger
- Keep paths filter on push to main (optimize deployments)
- Workflow now runs on ALL PRs to satisfy branch protection
Trade-off:
- PRs changing non-worker files will trigger unnecessary dry-run
- But this ensures required status check always reports
- Small cost for reliability and simpler configuration
Alternatives considered:
- Path-aware required checks: Not supported by GitHub
- Remove required check: Loses CI validation
- Add all paths to filter: Makes config brittle
- Add ANTHROPIC_SMALL_FAST_MODEL to all Kimi configuration templates
- Update installation scripts (npm, Unix, Windows) to include SMALL_FAST_MODEL
- Ensure Kimi API configuration matches official documentation format
- Bump version to 2.5.1 with updated changelog
Files updated:
- config/base-kimi.settings.json: Add SMALL_FAST_MODEL
- installers/install.sh: Update Unix template creation
- installers/install.ps1: Update PowerShell template creation
- scripts/postinstall.js: Update npm postinstall script
- All version files: Bump to 2.5.1
- Add kimi profile support alongside existing glm and default profiles
- Create base-kimi.settings.json configuration template
- Update all installation methods (npm, Unix, Windows) to auto-create Kimi settings
- Enhance documentation with Kimi examples and API setup instructions
- Update version to 2.5.0 with comprehensive changelog
- Add Kimi detection logic in install scripts for seamless migration
- Maintain backward compatibility with existing GLM and Claude profiles
Use string concatenation (not args array) when shell is needed to avoid
Node.js DEP0190 deprecation warning. Restores previous working approach
with conditional shell usage based on file extension.
Key changes:
- Added escapeShellArg() helper for proper argument escaping
- Conditional shell: only for .cmd/.bat/.ps1 files on Windows
- When shell needed: concatenate args into single escaped string
- When no shell: use array form (faster, no overhead)
Benefits:
- No deprecation warning
- Proper security (escaped arguments)
- Better performance (no shell on Unix or for .exe files)
- Matches previous stable implementation
Version: 2.4.9
Use platform-specific shell option (Windows only) instead of shell: true
to avoid deprecation warning in Node.js v22.9.0+.
Changes:
- bin/ccs.js: Change shell: true to shell: process.platform === 'win32'
- CHANGELOG.md: Document fix for v2.4.8
- VERSION: Bump to 2.4.8
- package.json: Bump to 2.4.8
Benefits:
- No deprecation warning on Windows
- Better performance on Unix (no shell overhead)
- Maintains Windows .cmd/.bat compatibility
- fix color detection for cross-platform TTY compatibility
- enhance help command with npm-specific content and npx examples
- remove --install/--uninstall flags pending .claude/ integration testing
- update version across all files and documentation
- preserve implementation code for future release readiness
BREAKING CHANGE: Reorganize tests into native/ and npm/ directories
- Move 37 native Unix tests to tests/native/unix/
- Move Windows tests to tests/native/windows/
- Move 39 npm package tests to tests/npm/
- Move shared utilities to tests/shared/
- Add comprehensive test documentation
- Implement master orchestrators for backward compatibility
- Increase test coverage from 41 to 83+ tests (100% increase)
- Add mocha framework for npm tests
- Clean up old test files and directory structure
New test commands:
- npm run test:npm (npm package tests only)
- npm run test:native (native tests only)
- npm run test:unit (unit tests only)
- npm run test:all (all tests)
Backward compatibility maintained:
- npm test (runs all tests)
- bash tests/edge-cases.sh (master orchestrator)
- All existing workflows unchanged
- Update .gitignore to exclude node_modules and include Node.js patterns
- Update package.json with final test scripts and structure
- Complete test restructure: 83 tests across native, npm, and unit suites
- All tests passing with 100% success rate
- Backward compatibility maintained
- Document final test organization with accurate test counts
- Update to reflect integration tests moved to npm/ directory
- Include detailed file listings and test counts
- Clarify backward compatibility and migration benefits
- Final counts: 37 native + 39 npm + 7 unit = 83 total tests
