Add project files, examples, and documentation

- Added package.json for npm/bun compatibility
- Added tsconfig.json for TypeScript configuration
- Added MIT LICENSE
- Added CHANGELOG.md for version tracking
- Added 4 example files demonstrating usage:
  - 01-basic-compaction.ts - Context compaction basics
  - 02-claude-code-integration.ts - Claude Code integration
  - 03-openclaw-integration.ts - OpenClaw integration
  - 04-state-machine-pipeline.ts - Deterministic pipeline

All extracted source files are included in the repository.
This commit is contained in:
admin
2026-03-03 13:18:04 +00:00
Unverified
parent c629646b9f
commit e6981ee8f8
8 changed files with 705 additions and 0 deletions

100
CHANGELOG.md Normal file
View File

@@ -0,0 +1,100 @@
# Changelog
All notable changes to this project will be documented in this file.
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
## [1.1.0] - 2025-03-03
### Added
- **Claude Code Integration** (`agent-system/integrations/claude-code.ts`)
- Full context compaction support for Claude Code CLI/IDE
- Subagent spawning with 6 predefined types
- Parallel subagent execution
- Persistent memory (remember/recall)
- Session management
- Tool registration for Claude API
- Auto-compaction on threshold
- **OpenClaw Integration** (`agent-system/integrations/openclaw.ts`)
- Deterministic pipeline creation
- Parallel agent execution (4 projects × 3 roles pattern)
- Workspace isolation with permissions and quotas
- Lobster YAML workflow parser
- Hook system for events
- Context save/restore across sessions
- **Examples** (`examples/`)
- Basic context compaction example
- Claude Code integration example
- OpenClaw integration example
- State machine pipeline example
- **Project Files**
- `package.json` for npm/bun compatibility
- `tsconfig.json` for TypeScript configuration
- `LICENSE` (MIT)
### Changed
- Updated `agent-system/index.ts` to export new integrations
- Updated README with comprehensive documentation and hero section
## [1.0.0] - 2025-03-03
### Added
- **Agent System** (`agent-system/`)
- `core/token-counter.ts` - Token counting and budget management
- `core/summarizer.ts` - LLM-powered conversation summarization
- `core/context-manager.ts` - Context compaction with 4 strategies
- `core/orchestrator.ts` - Agent lifecycle management
- `core/subagent-spawner.ts` - Subagent creation and execution
- `agents/base-agent.ts` - Base agent class
- `agents/task-agent.ts` - Task-specific agent
- `storage/memory-store.ts` - Persistent storage
- `utils/helpers.ts` - Utility functions
- **Pipeline System** (`pipeline-system/`)
- `core/state-machine.ts` - Deterministic state machine
- `engine/parallel-executor.ts` - Parallel execution engine
- `events/event-bus.ts` - Event-driven coordination
- `workspace/agent-workspace.ts` - Workspace isolation
- `workflows/yaml-workflow.ts` - YAML workflow parser
- `integrations/claude-code.ts` - Claude Code integration layer
- **Downloads** (`downloads/`)
- `agent-system.zip` - Agent system package
- `pipeline-system.zip` - Pipeline system package
- `complete-agent-pipeline-system.zip` - Full system package
- **Documentation**
- Comprehensive README with examples
- API reference tables
- Architecture diagrams
---
## Release Notes
### v1.1.0 - Integration Release
This release adds full integration support for both Claude Code and OpenClaw, making it easy to incorporate context compaction and deterministic pipeline orchestration into existing AI development workflows.
**Key Highlights:**
- One-line integration with Claude Code CLI
- OpenClaw-compatible pipeline definitions
- Support for the 4×3 parallel agent pattern (4 projects × 3 roles)
- Persistent memory across sessions
### v1.0.0 - Initial Release
Initial release of the Agentic Compaction & Pipeline System, providing:
- Token-aware context management
- 4 compaction strategies
- Deterministic state machine
- Parallel execution engine
- Event-driven coordination
---
*All releases are 100% autonomously built by Z.AI GLM-5*