fact-check-gate
Skill0 starsChanged 2 months agoMITView on GitHub
行途 XingTu 成稿"事实核查门禁"技能。把文章/观点/数据/方案里的说法拆成可验证事实、推论、 价值判断三部分,联网核查事实并打五档标记(已证实/需收窄/争议/证据不足/明显错误), 再查推理链漏洞,输出补强版与可信度结论——防 AI 幻觉、防编造第三方产品机制。 触发:事实核查、数字对不对、这个说法靠谱吗、核实数据、防编造、成稿前核查、 查证、这机制是真的吗、帮我把这段话核一遍。 依据:RULES.md PUB-018(脱敏+防编造红线)+ BCI B-009(编造第三方项目机制踩坑)+ xingtu-prompts/review/fact-check.md。 版本:v1.0(2026-09-11,豆包工作沉淀)。
1 other project uses a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- xingtu1996/dsh-xingtu-skillsedited, first seen Sep 7, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 1
- xingtu1996/dsh-xingtu-skills★ 295% same
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository139 other files it gives its agents.
- AGENTS.mdAGENTS.md
- skills/caveman-learn/CLAUDE.mdCLAUDE.md
- adversarial-reviewSkill
- ai-repo-engineering-layoutSkill
- ai-session-material-minerSkill
- ai-usage-reportSkill
- ardot-effect-fallbackSkill
- audio-friendly-adaptSkill
- automation-resurrectionSkill
- book2vido-videoSkill
- boss-autopilotSkill
- brand-icon-sourcingSkill
- canvas-designSkill
- cavecrewSkill
- caveman-commitSkill
- caveman-compressSkill
- caveman-discoverSkill
- caveman-evidence-reviewSkill
- caveman-exploreSkill
- caveman-helpSkill
- caveman-learnSkill
- caveman-manageSkill
- caveman-optimizeSkill
- caveman-reviewSkill
- caveman-setupSkill
- cavemanSkill
- caveman-statsSkill
- cc-switch-usage-auditSkill
- certification-submission-kitSkill
- comment-reply-playbookSkill
- concept-anchor-figureSkill
- conference-archiveSkill
- constitution-amendmentSkill
- content-format-auditSkill
- content-matrix-channel-auditSkill
- cross-repo-skill-curationSkill
- daily-judgment-briefSkill
- daily-reportSkill
- de-ai-flavorSkill
- desktop-ai-forensicsSkill
- discernment-nudgeSkill
- doc-translateSkill
- doubao-app-layer-archiveSkill
- effort-accountingSkill
- ego-browserSkill
- egress-allowlistSkill
- eval-skillsSkill
- evolve-skillsSkill
- exam-prep-pmpSkill
- fanout-mainline-ledgerSkill
- free-token-channel-auditSkill
- github-repo-gallerySkill
- git-identity-pinSkill
- git-push-in-sandboxSkill
- harness-governanceSkill
- harness-model-upgrade-auditSkill
- hidden-contract-auditSkill
- investigate-firstSkill
- ip-platform-fit-auditSkill
- lean-buildSkill
- The first 60 shown.
The file
Files
事实核查门禁(fact-check-gate · 行途 XingTu 版)
一句话
行途 XingTu 文章里的每个数字和机制描述都要经得起核查。 本技能把任何说法拆成"事实 / 推论 / 价值判断"三层,先核事实、再查推理,输出"可以相信到什么程度"——治 AI 幻觉,也治"凭印象硬编第三方机制"(BCI B-009)。
触发条件
- 用户对文章/观点/数据/方案说"事实核查""数字对不对""这个说法靠谱吗""核实一下""防编造""成稿前核查"
- 主动前置:成稿/排版前,文中含具体数字、第三方产品机制、引用结论时,自动跑一遍本技能(对齐 BCI B-009 防编造 Gate)
核查流程(五步)
Step 1:拆层
把待核查说法拆成三部分:
- 可验证的事实(能被外部证据证实/证伪)
- 从事实推出的结论(推理链)
- 价值判断("应该怎样"的偏好,不参与事实核查)
Step 2:事实核查(联网)
对每一条事实联网核查来源、样本、时间、完整上下文,打五档标记:
| 标记 | 含义 | 处理 |
| ① 已证实 | 一手来源/官方数据支撑 | 保留,标注来源与日期 |
| ② 基本成立,但需收窄 | 方向对、范围/条件需限定 | 收窄表述后保留 |
| ③ 存在争议 | 多方说法冲突 | 并列呈现,不选边 |
| ④ 证据不足 | 找不到可靠来源 | 明确写"暂未核实",不强写 |
| ⑤ 明显错误 | 与权威来源矛盾 | 删除或修正,标注错误点 |
Step 3:推理链检查
在"事实成立"的前提下继续查:
- 这些事实能否推出当前结论
- 是否藏着未经验证的假设
- 是否混淆相关性与因果性
- 是否遗漏其他解释或关键信息
- 结论在什么条件下成立/失效
Step 4:输出结论
按固定结构输出:
- 哪些事实可信、哪些需修正(逐条标注五档)
- 推理链中最关键的漏洞
- 补强后的最合理版本(可直接替换原文)
- 目前可以相信到什么程度
Step 5:落地处置
- ①/② → 保留,标来源日期(对齐 PUB-018 官方口径 vs 实测口径分开)
- ③/④ → 改稿为"暂未核实"或并列呈现,绝不硬写
- ⑤ → 直接修正,并记入 BCI(坏例索引)防复发
行途 XingTu 特化规则
| 场景 | 规则 |
| 第三方产品机制(BCI B-009) | 涉第三方项目的机制描述必须回一手源(GitHub API/官方 README/官方文档)核对;二手视频转述不得作为机制与数字依据;找不到官方说明就不写机制 |
| 数字口径(PUB-018) | 官方口径 vs 实测口径必须分开标注;对外数字一律脱敏到量级(公司/团队数字) |
| 多工具产物 | 批量成稿时每篇独立核查,不得"一篇查过=整批可信" |
| 自查卡/产品文案 | 承诺性表述同样过闸("能省 XX%"必须有实测或官方依据) |
反模式(要避免的)
- 只核事实不核推理:事实全对也可能推出错误结论(相关≠因果)
- 事实与观点混为一谈:没拆层直接下判断
- 用"大概率合理"代替核查:找不到证据就写"暂未核实",不靠推测补全
- 忽略价值判断:把"应该怎样"的偏好当成客观结论
- 一次性结论:五档标记只标"对/错"两级,丢失"需收窄/争议/证据不足"的中间态
用法示例
- 「帮我把这段话核一遍:某工具能省 60% token,官方基准是……」
- 「这篇文章里引用的 X 模型跑分对吗?来源是哪?什么时间点的数据?」
- 「这个新项目的机制描述和官方文档对得上吗?」(→ 回一手源核对,治 BCI B-009)
与其它技能联动
ai-usage-report:本地 cc-switch 数据核实(缓存命中率/倍率/token 账单),数字类核查优先走本地真实数据adversarial-review:对抗审查的"数据核查"角色可调用本技能做数字专项de-ai-flavor:去 AI 味时发现的存疑数字,交回本技能核查
原则:核查不是不写,是不写没有依据的东西。 数字可以少,但每个数字都要站得住。
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
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.

