Add ADR-004 storage layer with append-only executions, summary
regeneration, idempotency keys, and retention pruning. Wire memory init
to scaffold .kaizen/metrics/ by default and add unit tests.
Persist INTENT and ecosystem assessments in history/, add ADR-004 for
project metrics with Helix Forge correlation, and register WP-0003 and
WP-0004 workplans with State Hub. Update SCOPE, README, and agency-framework
docs to reflect the two-layer measurement model.
Add agents/agent-optimization.md — the Kaizen Optimizer meta-agent for
analyzing and improving agent performance. Also update .gitignore to
suppress agents_backup_*/ directories produced by optimization scripts.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
T25: add tests/test_e2e_agency_framework.py — 16 E2E tests covering the full
memory lifecycle (init, show, brief, clear) and protocol list/show commands.
T26: replace agency-framework.md protocols placeholder with full documentation —
location convention, frontmatter schema, CLI reference, sys-medic memory
extensions, and protocols table.
T27: add Related Documents footer to agent-sys-medic.md linking to the k3s
protocol runbook, ADR-002, ADR-003, and agency-framework.md.
Fix: rename CLI command function list() → list_agents() to stop it shadowing
Python's built-in list(). The shadow caused memory_brief() to invoke the
agent-list command instead of constructing a list from dict keys, producing
the agent list as output on every `memory brief` invocation.
All 27 WP-0002 tasks complete. Test suite: 51 passed, 1 skipped.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add agents/agent-coach.md: new meta-category coaching agent that reads
all project agent memories, synthesises cross-agent briefs, and produces
targeted orientation briefs for incoming agents
- Complete memory brief command: now reads all .kaizen/agents/*/memory.md,
formats structured orientation output following coach agent spec, adds
--raw flag for unformatted dump
- Coach validates and appears under kaizen-agentic list --category meta
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds a State Hub Task IDs table to each workplan file so task UUIDs are
always available without requiring list_tasks lookups in future sessions.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Integrates sys-medic as a standard kaizen-agentic agent with YAML frontmatter,
source attribution, and single-prompt format. Validated via list and validate.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Three-part workplan (27 tasks) covering:
- Part 1: sys-medic integration as standard kaizen-agentic agent (T01-T06)
- Part 2: agency framework — project memory model, coaching meta-agent,
and CLI memory command group (T07-T16)
- Part 3: sys-medic extended with protocols runbook and node-profile
memory, built on the Part 2 framework (T17-T27)
Workstream registered in state-hub as kaizen-wp-0002-agency-framework.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Moves 8 tasks from TODO.md [Unreleased] into
workplans/kaizen-agentic-WP-0001-community-engagement.md and registers
them in the state-hub as workstream kaizen-wp-0001-community-engagement.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Register with state-hub under custodian domain. Replace monolithic CLAUDE.md
with thin @-import index; distribute content into .claude/rules/ per ops-bridge pattern.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Using the repo in the experimental capability subrepo style breaks
coordination and we need to get the pieces together again. I will be
switching to using Pull Requests in the future.
- Extend safe_cli_wrapper() to suppress spurious Click errors for both
install and update commands; add success indicators for update output
- Add test_update_command_error_suppression to verify error suppression
- Expand CLAUDE.md to document all 17 agents with categories
- Add Keep a Contributing-File format header to CONTRIBUTING.md
- Fix TodoFileGuide URL reference in TODO.md
- Add RELEASE_NOTES_v1.0.1.md
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Update agent to reference TODO.md instead of NEXT.md for task management
following the todofile system implementation in main project.
- Updated key project files documentation to reference TODO.md
- Modified session wrap-up protocol to use todofile format
- Maintains consistency with main project's todofile adoption
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
🔧 Extended Click library workaround to cover update command
📚 Updated CLAUDE.md with all 17 agents organized by category
🧪 Added comprehensive test coverage for update command
✅ All 11 CLI error handling tests pass
Major improvements:
- Both install and update commands now provide clean professional output
- Complete agent ecosystem documentation (17 agents vs previous 6)
- Unified error handling approach across affected CLI commands
- Enhanced test suite covering all CLI error scenarios
User experience: Clean output for kaizen-agentic install/update commands
Technical: Extended safe_cli_wrapper to handle ["install", "update"] commands
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add make install-global target with intelligent installation methods
- Support pipx (recommended), pip --user, and fallback options
- Resolve issue where kaizen-agentic was only available in local venv
- Update documentation to clearly explain 4 installation options:
- Development mode (local venv)
- Global installation (any directory)
- Local package testing (local venv)
- Future PyPI installation
- CLI now available globally from any directory after make install-global
Fixes the core issue where users couldn't access kaizen-agentic from other repos.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Copy agent-releaseManager.md to src/kaizen_agentic/data/agents/
- Ensure releaseManager agent is packaged and distributed with v1.0.0
- Complete v1.0.0 package now includes all 17 agents including release management
- Final package ready for PyPI publication with full feature set
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add release management system with agent-releaseManager and 6 release- targets
- Include install-local target for PyPI-equivalent local testing
- Document enhanced Makefile with structured release workflow
- Complete v1.0.0 feature documentation for production release
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add Option B: From Local Package installation instructions
- Include make install-local target in both README.md and GETTING_STARTED.md
- Provide clear path for testing PyPI-equivalent installation locally
- Maintains development, local package, and future PyPI installation options
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add make install-local target to install from locally built wheel package
- Provides equivalent to PyPI installation for testing before publication
- Includes comprehensive installation validation with import and CLI tests
- Shows package information and validates core functionality
- Perfect for testing the installation experience without PyPI dependency
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Apply black formatting to all Python files
- Fix various flake8 violations in agent system code
- Clean up imports and whitespace issues
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add agent-releaseManager.md with comprehensive publication workflow guidance
- Add 6 release- prefixed make targets for structured release process:
- release-check: Validate release readiness
- release-prepare: Build packages and prepare release
- release-test: Test publication via TestPyPI
- release-publish: Publish to production PyPI
- release-finalize: Post-release tasks (tags, GitHub releases)
- release-rollback: Emergency rollback procedures
- Update pyproject.toml version from 0.1.0 to 1.0.0 for consistency with CHANGELOG.md
- Update installation documentation in README.md and GETTING_STARTED.md
- Add current "from source" installation instructions
- Maintain "from PyPI" instructions for post-publication
- Framework now ready for v1.0.0 publication with complete release workflow
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Update CHANGELOG.md with comprehensive v1.0.0 release notes
- Document both Scenario 1 (greenfield) and Scenario 2 (existing project integration) achievements
- Move completed Scenario 2 tasks from active to completed sections in TODO.md
- Mark both major usage scenarios as production-ready with full feature coverage
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Major documentation improvements:
- Fixed all outdated agent names throughout README.md and GETTING_STARTED.md
- Updated agent references from old names (todo-keeper, changelog-keeper, setup-repository)
to new names (keepaTodofile, keepaChangelog, setupRepository)
- Created comprehensive HELLO_WORLD_TUTORIAL.md with step-by-step guide for new users
- Added prominent tutorial links in README.md and GETTING_STARTED.md for better discoverability
- Updated Python version requirements from 3.8 to 3.9 in project templates
- Enhanced first-time user experience with clear guidance and working examples
Documentation now provides:
✅ Working commands that match actual agent names
✅ Clear separation between new users and experienced users
✅ Complete tutorial from installation to running code
✅ Proper guidance for scenario one (greenfield projects)
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Updated test fixtures and expectations to match new agent naming:
- Fixed test registry to use correct agent names (setupRepository, keepaTodofile, keepaChangelog)
- Updated test assertions to expect new agent names instead of old ones
- Added missing category fields to test agent definitions
- Enhanced test registry with keepaChangelog agent for complete template testing
All 24 tests now pass:
- test_core.py: 6 tests ✅
- test_installer.py: 10 tests ✅
- test_registry.py: 8 tests ✅🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Added two key usage scenarios to TODO.md for framework development:
Scenario 1: Establish codebase from scratch
- Research current onboarding experience and setup targets
- Identify documentation gaps for new project creation
- Validate project initialization workflows
- Evaluate agent selection systems for new projects
Scenario 2: Integration with existing projects having agents
- Research detection of existing agent systems
- Design conflict resolution for overlapping functionality
- Create migration paths for outdated project agents
- Develop extension mechanisms for project-specific customizations
- Define integration patterns respecting existing structure
- Build tooling for safe agent system transitions
These exploration tasks will guide development of capabilities needed
to support both greenfield projects and integration scenarios.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Fixed YAML frontmatter errors in all agent files causing loading failures
- Added proper category fields to all 16 agents for correct classification
- Standardized agent names to match filenames for consistency
- Updated TODO.md to reflect completed agent system fixes
- Enhanced agents-install-cli make target with pip upgrade and user guidance
All agents now load properly without warnings and display in correct categories:
- Documentation (1), Code Quality (4), Project Management (4)
- Development Process (3), Infrastructure (1), Testing (3)
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Used todo-keeper agent to comprehensively update the todofile:
MAJOR COMPLETION RECOGNITION:
- Moved all originally planned features to COMPLETED section
- Documented comprehensive CLI tool with 8 commands
- Recognized 24 tests with 100% pass rate achievement
- Acknowledged complete documentation system (4 guides)
- Celebrated code quality compliance (PEP 8, flake8 clean)
NEW ROADMAP FOR V0.3.0:
- Pre-commit hooks and CI/CD automation
- Enhanced error handling and performance optimization
- Cross-platform compatibility testing
- Agent metrics and telemetry systems
- Security enhancements and integrity verification
CURRENT STATE ASSESSMENT:
- Project transformed from basic framework to production-ready system
- All core functionality implemented and tested
- Remaining tasks focus on production enhancements
- Ready for broader adoption and deployment
Follows Keep a Todofile V0.0.1 format with proper categorization
and clear roadmap for continued development.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Used changelog-keeper agent to document major transformations since v0.1.0:
COMPREHENSIVE DOCUMENTATION:
- Complete hybrid agent distribution system implementation
- CLI tool (kaizen-agentic) with full command interface
- Agent registry, installer, and project initialization functionality
- Project templates system and agent categorization
- Makefile target reorganization with consistent naming
- All flake8 fixes and test reliability improvements
- New documentation guides (GETTING_STARTED, AGENT_DISTRIBUTION, CLI_CHEAT_SHEET)
PROPER CATEGORIZATION:
- Added: 15+ major new features and capabilities
- Changed: 6 significant improvements and reorganizations
- Fixed: 7 critical bug fixes and compliance issues
SEMANTIC VERSIONING ALIGNMENT:
- Documents transformation from basic framework to production-ready distribution system
- Represents major version bump significance (toward 1.0.0)
- Follows Keep a Changelog format with user-focused descriptions
The changelog now accurately reflects the massive evolution of this project
into a comprehensive agent distribution system with CLI tools and package management.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
CODE STYLE FIXES:
- Fixed line length violations by breaking long lines appropriately
- Removed unused imports (os from installer.py and registry.py)
- Removed unused variables (package_name in _create_pyproject_toml)
- Fixed f-string usage (removed f-strings without placeholders)
- Fixed import organization and removed redundant imports
- Added missing newlines at end of files
SPECIFIC FIXES:
- cli.py: Split long option line, fixed f-string usage, added newline
- installer.py: Removed unused imports/variables, fixed line breaks, improved validation logic
- registry.py: Removed unused import/variable, broke long condition line
- test files: Removed unused imports, fixed long assertion lines, added newlines
All 24 tests still pass and flake8 now reports no violations.
Code is now compliant with PEP 8 style guidelines.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
MAKEFILE TARGET RENAMING:
- Renamed agent management targets to use agents- prefix for consistency
- list-agents → agents-list
- update-agents → agents-update
- validate-agents → agents-validate
- agent-status → agents-status
- install-agent-cli → agents-install-cli
- Updated all documentation to reflect new naming convention
TEST FIXES:
- Fixed backup directory collision in tests by adding microseconds and counter
- Improved dependency detection to be more precise and avoid false positives
- Updated dependency parsing to support YAML frontmatter dependencies
- Fixed test expectations to match actual validation behavior
- All 24 tests now passing
DEPENDENCY SYSTEM IMPROVEMENTS:
- Enhanced dependency extraction from YAML frontmatter (dependencies, depends_on, requires)
- More precise agent reference detection in content
- Better handling of explicit vs implicit dependencies
- Improved validation error reporting
This ensures consistent naming convention (setup-, standards-, agents-) across
all Makefile targets while fixing test reliability issues.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Complete implementation of the agent distribution framework including:
CORE INFRASTRUCTURE:
- AgentRegistry: Agent discovery, categorization, and dependency management
- AgentInstaller: Agent installation, updates, and removal with safety measures
- ProjectInitializer: Template-based project initialization with agent integration
- CLI Tool: Comprehensive kaizen-agentic command-line interface
DISTRIBUTION FEATURES:
- Python package distribution with console script entry point
- Agent categorization (project-management, development-process, code-quality, etc.)
- Project templates (python-basic, python-web, python-cli, python-data, comprehensive)
- Dependency resolution and validation
- Idempotent operations with backup and rollback support
CLI COMMANDS:
- kaizen-agentic init: Initialize new projects with agents
- kaizen-agentic install/update/remove: Manage agents in existing projects
- kaizen-agentic list/status/validate: Discovery and maintenance
- kaizen-agentic templates: Project template management
INTEGRATION & DOCUMENTATION:
- Makefile targets for agent management (list-agents, update-agents, etc.)
- Automatic Claude Code configuration updates (CLAUDE.md)
- Comprehensive documentation (GETTING_STARTED, AGENT_DISTRIBUTION, CLI_CHEAT_SHEET)
- Multi-language build system integration examples
- Complete test coverage for all components
PACKAGE STRUCTURE:
- Console script: kaizen-agentic command available globally
- Package data: All agents included for distribution
- Dependencies: click, pyyaml for CLI and parsing
- Testing: Comprehensive test suite for registry and installer
This enables sharing specialized AI agents across projects with easy installation,
updates, and management through both CLI and integrated Makefile targets.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- ContextManifest.md: URLs for recreating external context dependencies
- KeepaContributingfile.md: Keep a Contributing-File V0.0.1 format reference
- KeepaTodofile.md: Keep a Todofile V0.0.1 format reference
- PythonVibes.md: Python project best practices reference
These files preserve the external documentation standards used throughout
the project for reproducible context and consistent agent behavior.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Essential project files: .gitignore, pyproject.toml, README.md
- Documentation framework: CHANGELOG.md, CONTRIBUTING.md, TODO.md, CLAUDE.md
- 15 specialized agent definitions for comprehensive development workflow
- Core source code structure with optimization framework
- Testing infrastructure with example tests
- Proper Python package structure following PythonVibes standards
This establishes the complete foundation for the AI agent development framework
with agent-driven workflows, continuous improvement principles, and comprehensive
development infrastructure.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Renamed verify-setup to setup-verify for consistency with setup-* targets
- Grouped all standards targets into dedicated section with standards- prefix
- Renamed check-standards → standards-check, fix-standards → standards-fix, test-standards → standards-test
- Updated help text with clear Standards Compliance section
- Added visual section separators for better organization
- Removed duplicate targets and cleaned up structure
- Removed obsolete agent-repository-structure.md in favor of comprehensive agent-setupRepository.md
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
