agentleFS
gootfgootf/eacks/skills/eacks-s5-artifacts/SKILL.md

eacks-s5-artifacts

Skill2 starsChanged 3 months agoMITView on GitHub

EACKS S5 图整合守卫——Relation Matrix / Conflict Index / Boundary Matrix 三类产物格式定义、非破坏性合并规则(equivalent 门槛 + merge-decision-log + Canonical 仅组织层)、冲突结构化(四类标记 + 双保留)。Use when 把验证过的关系整合为知识图、生成三类合规产物、裁决合并决策或标记合并违规时。

The file

Files
SKILL.md

EACKS S5 图整合守卫

配套:eacks-execution(骨架,本 skill 是 S5 槽位实现)。本 skill 族(eacks-execution + 8 个阶段守卫)即 EACKS v1.3 的完整执行形态.工具层:onto_merger(合并引擎,方向反转适配见第五节)。

职责:把验证过的关系(S4 输出)整合为知识图,产出三类合规产物,并裁决/标记合并决策。源概念永不物理删除;Canonical 仅组织层抽象。

Guard 接口(与骨架 block-contracts S5 一致):

  • 输入:验证后的 Relation 表 + Concept/Correspondence 表(S2)+ 合并候选(onto_merger 或人工提出)
  • 输出:知识图(graphml/表形式)+ Relation Matrix + Conflict Index + Boundary Matrix + 合并合规报告

一、三类产物格式(核心契约)

1. Relation Matrix(关系矩阵)

| relation_id | source_ref | target_ref | type | status | strength | structurality | criticality | condition | scope | evidence_refs | provenance_refs |

字段约束(对应架构文档第七节交付物 5):

  • status:candidate / validated / disputed / unresolved / rejected(S4 裁决结果直入)
  • structurality:non_structural / structural(structural 晋升需 S7 通过后——S5 默认 non_structural,仅标记"候选晋升"状态)
  • condition / scope:与依据 Claim 字段一致(S4 Anti-Compression 已校验)
  • evidence_refs:EvidenceLink ID 列表(可追溯)
  • provenance_refs:来源链引用(可追溯)

2. Conflict Index(冲突索引)

| conflict_id | claim_a | claim_b | conflict_type | status | evidence | resolution_note |

conflict_type 四值(对应 S4 矛盾处理结果):

  • conditional_divergence:条件/语境差异(双保留,合法)
  • genuine_contradiction:确证冲突(双保留,不消解——硬原则 3)
  • measurement_difference:测量差异(引用 S2 measurement-corresponding 记录)
  • competing_theories:竞争理论(双保留,供 S6 高阶综合识别)

status:active / resolved(仅当一方被 S8 人工裁决或新证据消除,且须有 resolution_note 引用裁决记录)

3. Boundary Matrix(边界矩阵)

| boundary_id | subject | boundary_type | boundary_value | source_ref | status |

boundary_type:applicability(适用范围)/ extrapolation_limit(外推极限)/ condition_boundary(条件边界)/ level_boundary(层级边界)/ measurement_boundary(测量边界) status:active / superseded(须有 Supersedes 引用——防边界静默变更,S7 第 3 项检查对照基准)

二、非破坏性合并规则

  1. 合并门槛(引用 S2 判定,不重判):仅 Correspondence 类型 = equivalent 且证据充分(S2 四门槛全过)才允许合并
  2. 合并记录(merge-decision-log):每次合并写入
    | merge_id | concept_a | concept_b | correspondence_ref | evidence | decision | source_concepts_kept |
    source_concepts_kept:必须为 true(源概念永不物理删除,只建 Canonical 组织层节点)
  3. Canonical 节点规则:Canonical 仅组织层抽象(指向源概念的聚合),不替代、不删除、不改变源概念内容
  4. 合并违规标记:
    • 非 equivalent 类型被合并 → FAIL(引用 S2 Correspondence 表为证)
    • 合并无证据 → FAIL
    • 源概念被删除/覆盖 → FAIL(破坏性整合,最高违规)
    • 违规合并须回滚(回 S2/S4 层),不可带病进图

三、冲突结构化

  • S4 判定的四类冲突全部显式入图(双保留原则:冲突双方都是图节点,冲突以 Conflict Index 记录)
  • 禁止行为:把 genuine_contradiction 合并为单方结论(= Conflict Erasure,S7 第 7 项检查的图侧对照基准)
  • 禁止行为:把 conditional_divergence 强行统一(= 条件抹除)

四、onto_merger 接线(方向反转)

  • 合并引擎(候选层):onto_merger 的 align 流程可生成合并候选(调用方案见 eacks_om_adapter.py)
  • 决策层反转:其 merges 输出必须经过第二节门槛过滤——仅 equivalent + 证据充分才采纳;其余全部降级为 Correspondence 记录(保持分离)
  • 输出映射:onto_merger merges/mappings 表 → EACKS merge-decision-log / Correspondence 表(字段转换 + 补证据引用——onto_merger 的 merges 不带证据,每条合并决策必须补证据才能生效)

五、输出格式

# S5 图整合合规报告
- 输入:V 条 validated 关系 / C 对合并候选
- 图统计:N 节点 / E 边 / K 冲突(分类计数)
- 合并:M 条采纳(全部 equivalent+证据 ✓)/ X 条降级为 Correspondence(理由:类型非 equivalent / 证据不足)
- 违规标记:0(或列出 FAIL 项 + 回滚动作)
- 产物:Relation Matrix(V 行)/ Conflict Index(K 行)/ Boundary Matrix(B 行)已生成

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.