Document scoring and acceptance formulas
Make the scoring implementation auditable by writing the formulas for quality scoring, task scores, pheromones, candidate fusion, consensus, and S07/S09 pass gates. Constraint: The user asked to turn the scoring explanation into a document under the Chinese docs line. Rejected: Leaving formulas only in chat | future reviewers need a repo artifact linked from the docs index. Confidence: high Scope-risk: narrow Directive: Keep future score changes synchronized with SCORING_AND_ACCEPTANCE_FORMULAS.zh-CN.md and the acceptance scripts. Tested: rg formula/index scan; docs secret-pattern scan; git diff --check; .venv/bin/python -B -m unittest tests.test_model_io_report_audit. Not-tested: Did not rerun live S07 because this is a documentation-only formula extraction. Co-authored-by: OmX <omx@oh-my-codex.dev>
This commit is contained in:
@@ -25,6 +25,7 @@
|
||||
|
||||
- `docs/AGENT_SWARM_QUALITY_STANDARD.zh-CN.md`:Agent / 蜂群 Agent 质量标准、测试标准来源、S07 外部 GitHub 代码任务分配和接手机制。
|
||||
- `docs/ACADEMIC_STANDARD_TEST_REPORT.zh-CN.md`:S01-S09 测试场景、测试过程、测试结果、算法清单和马尔可夫过程判定。
|
||||
- `docs/SCORING_AND_ACCEPTANCE_FORMULAS.zh-CN.md`:分数实现、公式、变量含义、S07/S09 和最小闭环通过判定。
|
||||
- `docs/MODEL_AGNET_IO_REPORT.zh-CN.md`:每个模型 / Agnet 的任务、输入、输出、评分和交接过程。
|
||||
- `docs/AGNET_FRAMEWORK_INPUT_OUTPUT_LOGIC.zh-CN.md`:不带代码的框架逻辑说明,只从 Agnet 输入、输出、交接和收敛解释实现过程。
|
||||
- `docs/INDUSTRY_STANDARD_AGNET_TEST_PLAN.zh-CN.md`:行业参考框架到本项目标准的映射。
|
||||
|
||||
Reference in New Issue
Block a user