build-evidence-map
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseBuild Evidence Map
构建证据图谱
Turn one contested question into a portable decision artifact that shows what
supports the current position, what pushes against it, and what remains unknown.
Do not use a graph to decorate an answer that has not been sourced.
For a simple factual claim or a general fact-checking request, use a verification
workflow such as instead. Use this skill when the relationships
between evidence, intermediate claims, trade-offs, and missing facts matter.
doublecheck将一个有争议的问题转化为可移植的决策产物,展示哪些内容支持当前立场、哪些内容反对当前立场,以及哪些内容仍然未知。
不要用图谱装饰一个没有来源支撑的回答。
对于简单的事实性主张或一般性事实核查请求,请改用 之类的核验工作流。
当证据、中间主张、权衡取舍和缺失事实之间的关系很重要时,使用此 skill。
doublecheckWorkflow
工作流
-
Frame one decision. Write one falsifiable question and one provisional position. Narrow the question until a reader can identify what action or belief the map is testing.
-
Collect bounded source regions. Prefer direct observations and primary sources. Record the URL or absolute local path, publisher, publication date, retrieval date, section/page/line/timestamp locator, and a short checkable excerpt. Read references/evidence-ladder.md when source quality is disputed.
-
Atomize the reasoning. Create only four node types:
- : the single current verdict;
position - : an intermediate proposition;
claim - : a faithful statement of one source region;
evidence - : a specific missing fact that could change the verdict.
unknown
-
Type every edge. Use,
supports,contradicts, orqualifies. Add a plain-language note explaining why the source node bears on the target. Topical similarity is not support. Different scope, date, or population is not automatically a contradiction.missing -
Preserve counterevidence. Do not delete contrary evidence because the provisional verdict survives it. Represent scope differences withedges.
qualifies -
Express uncertainty structurally. Do not invent confidence percentages. Add an, narrow the position, or qualify a claim.
unknown -
Write UTF-8 JSON with asuffix. Follow references/map-schema.md. Keep IDs short, stable, and semantic.
.doubt.json -
Validate fail-closed. Resolverelative to this
scripts/validate.mjs, then run it with Node.js 18 or newer:SKILL.mdbashnode <skill-directory>/scripts/validate.mjs decision.doubt.jsonThe bundled validator uses only Node.js built-ins and does not require npm or network access. Fix every finding before reporting success. Only say the map is valid when the command exitsand prints0followed by a 64-character receipt. A file hash, node count, JSON parse, or manual schema review is not a Doubt receipt. If deterministic validation cannot run, report that block instead of inventing success.VALIDRender the validated map only when the user has already installed; do not install or execute a remote package implicitly:doubt-ai@0.8.0bashdoubt map decision.doubt.json --out decision.html -
Verify source snapshots only with explicit network permission. The following command retrieves each recorded HTTP(S) source and fails closed if an excerpt cannot be matched:bash
doubt verify decision.doubt.json \ --out decision.verified.doubt.jsonNever run this command implicitly. Local file verification does not use the network. Do not write aobject by hand or hide a mismatch.verification -
Inspect the deliverable. Confirm that the question, verdict, counterevidence, unknowns, edge notes, and exact source regions remain readable. Treat JSON as the canonical editable artifact; HTML is a shareable view.
-
界定一个决策。 写下一个可证伪的问题和一个暂定立场。 不断收窄问题,直到读者能够明确该图谱正在检验什么行动或信念。
-
收集边界明确的来源区域。 优先采用直接观察和第一手来源。 记录 URL 或绝对本地路径、发布者、发布日期、检索日期、章节/页码/行号/时间戳定位信息,以及一段简短且可核验的摘录。 当来源质量存在争议时,阅读 references/evidence-ladder.md。
-
将推理原子化。 仅创建四种节点类型:
- :唯一的当前结论;
position - :中间命题;
claim - :忠实呈现一个来源区域的陈述;
evidence - :可能改变结论的某一具体缺失事实。
unknown
-
为每条边设定类型。 使用、
supports、contradicts或qualifies。添加通俗易懂的说明,解释源节点为何会影响目标节点。 主题相似并不构成支持。范围、日期或研究总体不同,并不自动意味着矛盾。missing -
保留反证。 不要因为暂定结论仍然成立就删除相反证据。 使用边表示适用范围差异。
qualifies -
结构化地表达不确定性。 不要虚构置信度百分比。 添加、收窄立场或限定主张。
unknown -
编写采用 UTF-8 编码且带后缀的 JSON。 遵循 references/map-schema.md。保持 ID 简短、稳定且具有语义。
.doubt.json -
以 fail-closed 方式进行验证。 以本为基准解析
SKILL.md,然后使用 Node.js 18 或更高版本运行:scripts/validate.mjsbashnode <skill-directory>/scripts/validate.mjs decision.doubt.json捆绑的验证器仅使用 Node.js 内置模块,不需要 npm 或网络访问。 报告成功前,修复每一项问题。只有当命令退出码为且打印0后跟一个 64 字符的校验回执时,才能说明图谱有效。 文件哈希、节点数量、JSON 解析或手动 schema 审查均不能替代 Doubt 回执。 如果无法运行确定性验证,应报告这一阻碍,而不是捏造成功。VALID仅当用户已经安装时,才渲染已验证的图谱; 不要隐式安装或执行远程软件包:doubt-ai@0.8.0bashdoubt map decision.doubt.json --out decision.html -
仅在获得明确网络许可后核验来源快照。 以下命令会获取每个已记录的 HTTP(S) 来源;如果无法匹配摘录,则以 fail-closed 方式失败:bash
doubt verify decision.doubt.json \ --out decision.verified.doubt.json绝不隐式运行此命令。本地文件核验不使用网络。不要手动编写对象,也不要隐藏不匹配情况。verification -
检查交付产物。 确认问题、结论、反证、未知项、边说明和精确来源区域仍然清晰可读。 将 JSON 视为规范的可编辑产物;HTML 则是可共享视图。
Quality gates
质量门槛
A finished map must satisfy all of these:
- exactly one has incoming reasoning;
position - every evidence node names one source and participates in an edge;
- every source is used and has dates, a bounded locator, and a substantive excerpt;
- every non-position node has a directed path to the position;
- the reasoning graph has no duplicate edges or directed cycles;
- contrary or qualifying evidence is present when the source set contains it;
- each decision-changing gap is an explicit node;
unknown - every edge note explains support, contradiction, qualification, or absence;
- the verdict is no broader than the evidence.
完成的图谱必须满足以下所有条件:
- 恰好一个 节点拥有传入的推理关系;
position - 每个 节点都指定一个来源并参与至少一条边;
evidence - 每个来源均被使用,且具有日期、边界明确的定位信息和实质性摘录;
- 每个非 节点都有一条通向
position的有向路径;position - 推理图谱中不存在重复边或有向环;
- 来源集合中包含相反或限定性证据时,图谱中必须呈现这些证据;
- 每个可能改变决策的缺口都对应一个明确的 节点;
unknown - 每条边的说明都解释支持、矛盾、限定或缺失关系;
- 结论的宽泛程度不得超过证据所能支持的范围。
Deliver the result
交付结果
Report:
- the current position in one sentence;
- the strongest counterevidence or qualification;
- the most important unresolved unknown;
- paths to the canonical JSON and any rendered HTML;
- whether deterministic validation and explicit source verification ran.
Never describe a structurally valid map as proven true. Validation establishes
traceability and graph integrity; source quality and inference quality still
require human review.
报告:
- 用一句话说明当前立场;
- 最有力的反证或限定条件;
- 最重要的未解决未知项;
- 规范 JSON 及任何已渲染 HTML 的路径;
- 是否运行了确定性验证和明确的来源核验。
绝不能把结构上有效的图谱描述为已被证明正确。
验证只能确立可追溯性和图谱完整性;来源质量和推理质量仍需人工审查。