openclaw-alignment 1.0.0__tar.gz

openclaw_alignment-1.0.0/CHANGELOG.md

@@ -0,0 +1,60 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on Keep a Changelog and this project follows Semantic Versioning.
+
+## [1.0.0] - 2026-03-01
+
+### Added
+
+- Packaging metadata and entrypoints via `pyproject.toml`, including `openclaw-alignment` CLI.
+- Cross-platform installer scripts:
+  - `scripts/install.py`
+  - `scripts/install_unix.sh`
+  - `scripts/install_windows.ps1`
+- Release governance and community docs:
+  - `CONTRIBUTING*.md`
+  - `SECURITY*.md`
+  - `SUPPORT*.md`
+  - `CODE_OF_CONDUCT*.md`
+  - `RELEASING*.md`
+- CI/CD workflow templates:
+  - cross-platform matrix CI
+  - package build validation
+  - release publishing workflow
+  - dependency security audit workflow
+- Path abstraction utilities for cross-platform config/cache/state directories (`lib/paths.py`).
+- Audit artifacts and release scope docs under `audit/` and `docs/release/`.
+
+### Changed
+
+- Stabilized environment input normalization:
+  - invalid `task_type` falls back safely
+  - invalid/out-of-range `time_of_day` is normalized
+  - explicit precondition error when `step()` is called before `reset()`
+- Deterministic policy behavior for inference (`explore=False` now uses greedy selection).
+- Improved config update behavior to preserve wrapper schema and create parent dirs automatically.
+- Unified state/action dimension contracts through `lib/contracts.py` and consistent references.
+- Updated README/README_EN to match actual install paths, dependencies, and support matrix.
+
+### Fixed
+
+- CLI parser option bug that previously caused `python -m lib.integration --help` to fail.
+- Action decoding now returns explicit validation errors for malformed/overflow indices.
+- Docs consistency checker now handles both pytest collection output formats.
+- Removed generated model artifacts from source tracking to avoid dirty-tree churn and release noise.
+
+### Security
+
+- Added private vulnerability reporting policy and response SLO documentation.
+- Added scheduled dependency audit workflow (`pip-audit`).
+
+### Breaking Changes
+
+- None.
+
+### Known Limitations
+
+- Phase3 full dependency installation is heavier (includes optional ML/monitoring stack).
+- Legacy OpenClaw config/model paths are still supported for backward compatibility.

openclaw_alignment-1.0.0/CODE_OF_CONDUCT.md

@@ -0,0 +1,14 @@
+# Code of Conduct
+
+We are committed to a welcoming and harassment-free community.
+
+## Our Standards
+
+- Be respectful and constructive.
+- Assume good intent and discuss technical details in good faith.
+- No harassment, abuse, discrimination, or doxxing.
+
+## Enforcement
+
+Report incidents to project maintainers via private channels listed in `SECURITY.md`.
+Maintainers may remove content or contributors violating this code.

openclaw_alignment-1.0.0/CODE_OF_CONDUCT.zh-CN.md

@@ -0,0 +1,14 @@
+# 行为准则
+
+我们致力于建设友好、无骚扰的社区环境。
+
+## 社区标准
+
+- 保持尊重、建设性沟通。
+- 默认善意，围绕技术事实讨论。
+- 禁止骚扰、辱骂、歧视、人肉等行为。
+
+## 执行
+
+如有违规，请通过 `SECURITY.zh-CN.md` 中的私密渠道联系维护者。
+维护者有权删除违规内容或限制相关账号参与。

openclaw_alignment-1.0.0/CONTRIBUTING.md

@@ -0,0 +1,42 @@
+# Contributing
+
+Thanks for contributing to OpenClaw Alignment.
+
+## Development Setup
+
+```bash
+git clone https://github.com/412984588/openclaw-alignment.git
+cd openclaw-alignment
+python3 scripts/install.py --dev --editable
+```
+
+## Workflow
+
+1. Create a feature branch.
+2. Write/adjust tests first for behavior changes.
+3. Keep commits focused and reviewable.
+4. Run checks before opening PR.
+
+## Local Checks
+
+```bash
+python3 -m pytest tests/ -v
+python3 scripts/check_docs_consistency.py
+python3 -m build
+```
+
+## Pull Requests
+
+- Explain problem, solution, and test evidence.
+- Link issue if available.
+- Update docs/changelog for user-visible changes.
+
+## Commit Style
+
+Use Conventional Commits:
+
+- `feat: ...`
+- `fix: ...`
+- `docs: ...`
+- `chore: ...`
+- `test: ...`

openclaw_alignment-1.0.0/CONTRIBUTING.zh-CN.md

@@ -0,0 +1,42 @@
+# 贡献指南
+
+感谢你为 OpenClaw Alignment 贡献代码。
+
+## 开发环境
+
+```bash
+git clone https://github.com/412984588/openclaw-alignment.git
+cd openclaw-alignment
+python3 scripts/install.py --dev --editable
+```
+
+## 工作流
+
+1. 从主分支切功能分支。
+2. 行为改动优先补测试。
+3. 提交保持小而清晰。
+4. 提交 PR 前完成本地检查。
+
+## 本地检查
+
+```bash
+python3 -m pytest tests/ -v
+python3 scripts/check_docs_consistency.py
+python3 -m build
+```
+
+## PR 要求
+
+- 说明问题、方案与测试结果。
+- 有 issue 则关联 issue。
+- 用户可见变更需同步文档与 changelog。
+
+## 提交信息规范
+
+建议使用 Conventional Commits：
+
+- `feat: ...`
+- `fix: ...`
+- `docs: ...`
+- `chore: ...`
+- `test: ...`

openclaw_alignment-1.0.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 OpenClaw Community
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

openclaw_alignment-1.0.0/MANIFEST.in

@@ -0,0 +1,22 @@
+include README.md
+include README_EN.md
+include LICENSE
+include CHANGELOG.md
+include CONTRIBUTING.md
+include CONTRIBUTING.zh-CN.md
+include SECURITY.md
+include SECURITY.zh-CN.md
+include SUPPORT.md
+include SUPPORT.zh-CN.md
+include CODE_OF_CONDUCT.md
+include CODE_OF_CONDUCT.zh-CN.md
+include RELEASING.md
+include RELEASING.zh-CN.md
+include requirements.txt
+include requirements-full.txt
+include requirements-dev.txt
+recursive-include config *.json
+recursive-include skills *.md
+recursive-include audit *.md
+recursive-include docs *.md
+recursive-include scripts *.py *.sh *.ps1

openclaw_alignment-1.0.0/PKG-INFO

@@ -0,0 +1,144 @@
+Metadata-Version: 2.4
+Name: openclaw-alignment
+Version: 1.0.0
+Summary: Reinforcement-learning driven alignment engine for OpenClaw workflows
+Author: OpenClaw Alignment Contributors
+License-Expression: MIT
+Project-URL: Homepage, https://github.com/412984588/openclaw-alignment
+Project-URL: Repository, https://github.com/412984588/openclaw-alignment
+Project-URL: Issues, https://github.com/412984588/openclaw-alignment/issues
+Keywords: reinforcement-learning,alignment,openclaw,actor-critic
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Operating System :: Microsoft :: Windows
+Classifier: Operating System :: MacOS
+Classifier: Operating System :: POSIX :: Linux
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy>=1.24
+Provides-Extra: phase3
+Requires-Dist: redis>=5.0; extra == "phase3"
+Requires-Dist: celery>=5.3; extra == "phase3"
+Requires-Dist: torch>=2.2; extra == "phase3"
+Requires-Dist: tensorboard>=2.15; extra == "phase3"
+Requires-Dist: numba>=0.59; extra == "phase3"
+Requires-Dist: scipy>=1.11; extra == "phase3"
+Provides-Extra: dev
+Requires-Dist: pytest>=8.0; extra == "dev"
+Requires-Dist: pytest-cov>=5.0; extra == "dev"
+Requires-Dist: build>=1.2; extra == "dev"
+Requires-Dist: twine>=5.0; extra == "dev"
+Requires-Dist: pip-audit>=2.7; extra == "dev"
+Requires-Dist: ruff>=0.6; extra == "dev"
+Requires-Dist: mypy>=1.10; extra == "dev"
+Dynamic: license-file
+
+# OpenClaw Alignment System
+
+> Reinforcement-learning driven workflow alignment engine (Actor-Critic)
+
+**English** | **[简体中文](README.md)**
+
+## Features
+
+- Actor-Critic RL core pipeline
+- Four-dimensional reward system (objective/behavior/explicit/pattern)
+- Optional Phase3 modules (distributed training, tuning, monitoring, performance)
+- Contract drift guards (state/action dimensions + docs consistency)
+- Cross-platform support: Windows / macOS / Linux
+
+## Support Matrix
+
+- Python: 3.10, 3.11, 3.12, 3.13
+- OS: Windows, macOS, Linux
+
+## Installation
+
+### 1) PyPI (Recommended)
+
+```bash
+pip install openclaw-alignment
+```
+
+Optional Phase3 extras:
+
+```bash
+pip install "openclaw-alignment[phase3]"
+```
+
+### 2) Install from source
+
+```bash
+git clone https://github.com/412984588/openclaw-alignment.git
+cd openclaw-alignment
+python3 scripts/install.py
+```
+
+Development install:
+
+```bash
+python3 scripts/install.py --dev --editable
+```
+
+## Quick Verification
+
+```bash
+python3 -m pytest tests/ -v
+python3 scripts/check_docs_consistency.py
+openclaw-alignment --help
+```
+
+## Architecture
+
+### Core (Phase 1-2)
+
+- `lib/reward.py`: reward calculation engine
+- `lib/environment.py`: interaction environment
+  - `State`: State data class (17 dimensions)
+  - `Action`: Action data class (11 dimensions)
+- `lib/agent.py`: Actor-Critic agent
+- `lib/learner.py`: online learner
+- `lib/trainer.py`: training loop
+- `lib/contracts.py`: single source of truth for dimensions
+
+### Optional (Phase 3)
+
+- `lib/distributed_trainer.py`
+- `lib/hyperparameter_tuner.py`
+- `lib/monitoring.py`
+- `lib/performance_optimizer.py`
+
+## Documentation
+
+- Architecture: `docs/architecture.md`
+- Reward model: `docs/reward-model.md`
+- Configuration: `docs/configuration.md`
+- Optional dependencies: `docs/phase3-optional-deps.md`
+- Contributing: `CONTRIBUTING.md`
+- Security: `SECURITY.md`
+- Support: `SUPPORT.md`
+
+## Test Coverage
+
+- **Total Tests**: 80
+- **Pass Rate**: 100%
+- **Core RL + integration**: 54 tests ✅
+- **Phase 2**: 1 test ✅
+- **Phase 3**: 21 tests ✅
+- **Docs/contract drift guards**: 4 tests ✅
+
+## Release and Versioning
+
+- Versioning: SemVer (stable branch: `release/1.0.x`)
+- Release runbook: `RELEASING.md` / `RELEASING.zh-CN.md`
+- Changelog: `CHANGELOG.md`
+
+## License
+
+MIT

openclaw_alignment-1.0.0/README.md

@@ -0,0 +1,103 @@
+# OpenClaw 意图对齐系统
+
+> 强化学习驱动的工作流对齐引擎（Actor-Critic）
+
+**[English](README_EN.md)** | **简体中文**
+
+## 特性
+
+- Actor-Critic 强化学习主链路
+- 四维度奖励系统（客观指标/行为信号/显性反馈/模式偏好）
+- 可选 Phase3（分布式训练、自动调参、监控、性能优化）
+- 契约防漂移检查（状态/动作维度与文档一致性）
+- 跨平台支持：Windows / macOS / Linux
+
+## 支持矩阵
+
+- Python: 3.10, 3.11, 3.12, 3.13
+- OS: Windows, macOS, Linux
+
+## 安装
+
+### 1) PyPI（推荐）
+
+```bash
+pip install openclaw-alignment
+```
+
+可选 Phase3 依赖：
+
+```bash
+pip install "openclaw-alignment[phase3]"
+```
+
+### 2) 源码安装
+
+```bash
+git clone https://github.com/412984588/openclaw-alignment.git
+cd openclaw-alignment
+python3 scripts/install.py
+```
+
+开发安装：
+
+```bash
+python3 scripts/install.py --dev --editable
+```
+
+## 快速验证
+
+```bash
+python3 -m pytest tests/ -v
+python3 scripts/check_docs_consistency.py
+openclaw-alignment --help
+```
+
+## 架构
+
+### Core（Phase 1-2）
+
+- `lib/reward.py`: 四维度奖励计算
+- `lib/environment.py`: 交互环境
+  - State: 状态数据类（17维）
+  - Action: 动作数据类（11维）
+- `lib/agent.py`: Actor-Critic 智能体
+- `lib/learner.py`: 在线学习器
+- `lib/trainer.py`: 训练循环
+- `lib/contracts.py`: 维度契约唯一来源
+
+### Optional（Phase 3）
+
+- `lib/distributed_trainer.py`
+- `lib/hyperparameter_tuner.py`
+- `lib/monitoring.py`
+- `lib/performance_optimizer.py`
+
+## 文档
+
+- 架构：`docs/architecture.zh-CN.md`
+- 奖励模型：`docs/reward-model.zh-CN.md`
+- 配置：`docs/configuration.zh-CN.md`
+- 可选依赖：`docs/phase3-optional-deps.zh-CN.md`
+- 贡献：`CONTRIBUTING.zh-CN.md`
+- 安全：`SECURITY.zh-CN.md`
+- 支持：`SUPPORT.zh-CN.md`
+
+## 测试覆盖
+
+- **总测试数**: 80个
+- **通过率**: 100%
+- **核心RL与集成**: 54个测试 ✅
+- **Phase 2**: 1个测试 ✅
+- **Phase 3**: 21个测试 ✅
+- **文档/契约防漂移**: 4个测试 ✅
+
+## 发布与版本
+
+- 版本策略：SemVer（稳定分支：`release/1.0.x`）
+- 发布流程：见 `RELEASING.zh-CN.md` / `RELEASING.md`
+- 变更记录：`CHANGELOG.md`
+
+## 许可证
+
+MIT

openclaw_alignment-1.0.0/README_EN.md

@@ -0,0 +1,103 @@
+# OpenClaw Alignment System
+
+> Reinforcement-learning driven workflow alignment engine (Actor-Critic)
+
+**English** | **[简体中文](README.md)**
+
+## Features
+
+- Actor-Critic RL core pipeline
+- Four-dimensional reward system (objective/behavior/explicit/pattern)
+- Optional Phase3 modules (distributed training, tuning, monitoring, performance)
+- Contract drift guards (state/action dimensions + docs consistency)
+- Cross-platform support: Windows / macOS / Linux
+
+## Support Matrix
+
+- Python: 3.10, 3.11, 3.12, 3.13
+- OS: Windows, macOS, Linux
+
+## Installation
+
+### 1) PyPI (Recommended)
+
+```bash
+pip install openclaw-alignment
+```
+
+Optional Phase3 extras:
+
+```bash
+pip install "openclaw-alignment[phase3]"
+```
+
+### 2) Install from source
+
+```bash
+git clone https://github.com/412984588/openclaw-alignment.git
+cd openclaw-alignment
+python3 scripts/install.py
+```
+
+Development install:
+
+```bash
+python3 scripts/install.py --dev --editable
+```
+
+## Quick Verification
+
+```bash
+python3 -m pytest tests/ -v
+python3 scripts/check_docs_consistency.py
+openclaw-alignment --help
+```
+
+## Architecture
+
+### Core (Phase 1-2)
+
+- `lib/reward.py`: reward calculation engine
+- `lib/environment.py`: interaction environment
+  - `State`: State data class (17 dimensions)
+  - `Action`: Action data class (11 dimensions)
+- `lib/agent.py`: Actor-Critic agent
+- `lib/learner.py`: online learner
+- `lib/trainer.py`: training loop
+- `lib/contracts.py`: single source of truth for dimensions
+
+### Optional (Phase 3)
+
+- `lib/distributed_trainer.py`
+- `lib/hyperparameter_tuner.py`
+- `lib/monitoring.py`
+- `lib/performance_optimizer.py`
+
+## Documentation
+
+- Architecture: `docs/architecture.md`
+- Reward model: `docs/reward-model.md`
+- Configuration: `docs/configuration.md`
+- Optional dependencies: `docs/phase3-optional-deps.md`
+- Contributing: `CONTRIBUTING.md`
+- Security: `SECURITY.md`
+- Support: `SUPPORT.md`
+
+## Test Coverage
+
+- **Total Tests**: 80
+- **Pass Rate**: 100%
+- **Core RL + integration**: 54 tests ✅
+- **Phase 2**: 1 test ✅
+- **Phase 3**: 21 tests ✅
+- **Docs/contract drift guards**: 4 tests ✅
+
+## Release and Versioning
+
+- Versioning: SemVer (stable branch: `release/1.0.x`)
+- Release runbook: `RELEASING.md` / `RELEASING.zh-CN.md`
+- Changelog: `CHANGELOG.md`
+
+## License
+
+MIT

openclaw_alignment-1.0.0/RELEASING.md

@@ -0,0 +1,27 @@
+# Releasing
+
+## Prerequisites
+
+- CI is green on all supported platforms.
+- `python3 -m pytest tests/ -v` passes.
+- `python3 scripts/check_docs_consistency.py` passes.
+- `CHANGELOG.md` updated.
+
+## Release Steps
+
+1. Bump version in `pyproject.toml` and changelog.
+2. Create release branch/tag:
+   - `git tag vX.Y.Z`
+   - `git push origin vX.Y.Z`
+3. Build package:
+   - `python3 -m build`
+4. Validate artifacts:
+   - `python3 -m twine check dist/*`
+5. Publish:
+   - TestPyPI first, then PyPI.
+6. Create GitHub Release and paste changelog summary.
+
+## Post-release
+
+- Monitor issues for 7 days.
+- Prepare `X.Y.(Z+1)` hotfix if needed.

openclaw_alignment-1.0.0/RELEASING.zh-CN.md

@@ -0,0 +1,27 @@
+# 发布流程
+
+## 前置条件
+
+- CI 在所有支持平台通过。
+- `python3 -m pytest tests/ -v` 全绿。
+- `python3 scripts/check_docs_consistency.py` 通过。
+- `CHANGELOG.md` 已更新。
+
+## 发布步骤
+
+1. 更新 `pyproject.toml` 与 changelog 版本号。
+2. 打 tag 并推送：
+   - `git tag vX.Y.Z`
+   - `git push origin vX.Y.Z`
+3. 构建包：
+   - `python3 -m build`
+4. 校验包：
+   - `python3 -m twine check dist/*`
+5. 发布：
+   - 先 TestPyPI，再正式 PyPI。
+6. 创建 GitHub Release 并附变更摘要。
+
+## 发布后
+
+- 首周跟踪 issue。
+- 必要时发布 `X.Y.(Z+1)` 热修版本。

openclaw_alignment-1.0.0/SECURITY.md

@@ -0,0 +1,23 @@
+# Security Policy
+
+## Supported Versions
+
+- `1.0.x`: supported
+- `<1.0.0`: best effort only
+
+## Reporting a Vulnerability
+
+Please report vulnerabilities privately:
+
+- GitHub private advisory: https://github.com/412984588/openclaw-alignment/security/advisories/new
+- Security email: security@openclaw-alignment.example
+
+Before public release, maintainers must replace the placeholder email with the real security contact.
+
+Do **not** open public issues for unpatched vulnerabilities.
+
+## Response Targets
+
+- Initial response: within 72 hours
+- Triage decision: within 7 days
+- Patch target: based on severity and impact

openclaw_alignment-1.0.0/SECURITY.zh-CN.md

@@ -0,0 +1,23 @@
+# 安全策略
+
+## 支持版本
+
+- `1.0.x`：官方支持
+- `<1.0.0`：尽力支持
+
+## 漏洞上报
+
+请使用私密方式上报：
+
+- GitHub 私有安全通告：https://github.com/412984588/openclaw-alignment/security/advisories/new
+- 安全邮箱：security@openclaw-alignment.example
+
+发布前请将上述占位邮箱替换为真实安全联系方式。
+
+**不要**在公开 issue 中披露未修复漏洞。
+
+## 响应目标
+
+- 首次响应：72 小时内
+- 分级结论：7 天内
+- 修复时限：按严重级别与影响范围确定

openclaw_alignment-1.0.0/SUPPORT.md

@@ -0,0 +1,14 @@
+# Support
+
+## Where to Ask Questions
+
+- Usage questions: GitHub Discussions
+- Bug reports: GitHub Issues
+- Security issues: see `SECURITY.md`
+
+## What to Include
+
+- OS and Python version
+- Install method (`pip`/source)
+- Minimal reproduction steps
+- Error logs/traceback