Metadata-Version: 2.4
Name: self-improving-loop
Version: 0.1.3
Summary: A hexagram-guided reliability loop for AI agents: trace runtime signals, choose guarded patches, and roll back regressions.
Author-email: TaijiOS maintainers <yangfei222666-9@users.noreply.github.com>
License-Expression: MIT
Project-URL: Homepage, https://github.com/yangfei222666-9/self-improving-loop
Project-URL: Repository, https://github.com/yangfei222666-9/self-improving-loop
Project-URL: Issues, https://github.com/yangfei222666-9/self-improving-loop/issues
Project-URL: Parent Project, https://github.com/yangfei222666-9/taiji
Keywords: ai-agent,ai-agents,self-improving,feedback-loop,auto-rollback,rollback,adaptive-threshold,agent-evolution,llm-tooling,autonomous-agents,taijios
Classifier: Development Status :: 7 - Inactive
Classifier: Intended Audience :: Developers
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Provides-Extra: telegram
Requires-Dist: requests>=2.28; extra == "telegram"
Provides-Extra: dev
Requires-Dist: pytest>=7.0; extra == "dev"
Requires-Dist: pytest-cov>=4.0; extra == "dev"
Requires-Dist: black<26,>=25.1; extra == "dev"
Requires-Dist: ruff>=0.15; extra == "dev"
Requires-Dist: mypy<1.20,>=1.19; extra == "dev"
Requires-Dist: pre-commit>=4.0; extra == "dev"
Requires-Dist: build>=1.0; extra == "dev"
Requires-Dist: twine>=6.0; extra == "dev"
Dynamic: license-file

> ⚠️ **迁移说明(2026-08)**:本仓库是独立概念件(rollback-first reliability layer),自 2026-05 起停更;主线已迁移至 [taiji](https://github.com/yangfei222666-9/taiji)(证据门运行时,含 false-pass gate 与 CI 门禁)。概念仍有效,欢迎对照阅读。

> ⚠️ **Moved (2026-08)**: this repository is a standalone concept piece (a rollback-first reliability layer) and has not been updated since 2026-05. The mainline now lives in [taiji](https://github.com/yangfei222666-9/taiji), an evidence-gate runtime with the false-pass gate and CI gates. The concepts still apply and are worth reading alongside it.

# self-improving-loop

> A regression guard for AI agents.
