What is observe-infer?

A reasoning framework for inferring an unseen internal condition from visible or measurable signs. It emphasizes comparing several signals instead of relying on one observation.

Use it to connect observable behavior, measurements, or symptoms with possible underlying causes. It is suited to pattern analysis and indirect assessment when direct measurement is unavailable.

It provides a structured way to investigate systems, people, or situations whose internal state cannot be directly inspected, while reducing guesses based on a single clue.

observe-infer — Skill for Claude Code by dhicoc

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add dhicoc/wuyun-liuqi-skills --skill observe-infer

Clone the repo

git clone --depth 1 https://github.com/dhicoc/wuyun-liuqi-skills

Made for: Claude Code.

Or install wuyun-liuqi-skills, the plugin that ships this one along with the rest of its 38 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/dhicoc/wuyun-liuqi-skills/observe-infer/github.svg)](https://agentmods.dev/skills/dhicoc/wuyun-liuqi-skills/observe-infer)

Your own site

<a href="https://agentmods.dev/skills/dhicoc/wuyun-liuqi-skills/observe-infer"><img src="https://agentmods.dev/badge/skills/dhicoc/wuyun-liuqi-skills/observe-infer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/dhicoc/wuyun-liuqi-skills/observe-infer"><img src="https://agentmods.dev/badge/skills/dhicoc/wuyun-liuqi-skills/observe-infer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 74 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,691 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00074	$0.01691
Opus 5	$0.00037	$0.00846
Sonnet 5	$0.00015	$0.00338
Haiku 4.5	$0.00007	$0.00169

Measured 12d ago against content hash 49bae7bb215c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

observe-infer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

scripts/lib/neijing_snapshot/lingshu/observe-infer/SKILL.md · 123 lines

How it starts

The opening of the file, as written. The whole thing — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.

司外揣内诊断术

R — 原文 (Reading)

远者司外揣内, 近者司内揣外, 是谓阴阳之极, 天地之盖。视其外应, 以知其内藏, 则知所病矣。合而察之, 切而验之, 见而得之, 若清水明镜之不失其形也。

— 黄帝/岐伯, 外揣第四十五、本藏第四十七

I — 方法论骨架 (Interpretation)

司外揣内是一个从可观测外部信号推断不可直接观测的内部状态的结构化推理框架。

其核心操作分为四步:

建立映射: 确定外部可观测信号与内部隐藏状态之间的对应关系。面色对应五脏, 脉象对应气血——每种外部信号都是某个内部状态的"代理指标"。
多信号采集: 不能只看一个指标。灵枢强调"合而察之, 切而验之"——面色+脉象+体态+尺肤多信号交叉验证, 类似"多摄像头监控"。
差异对比: 通过对比不同位置的信号差异(如左vs右、上vs下)来精确定位问题。"独热""独寒""独大""独小"——异常总是从差异中被发现。
推断内部: 从外部信号的模式推断内部状态, 类似"清水明镜之不失其形"——信号越清晰, 推断越准确。

A1 — 书中的应用 (Past Application)

案例 1: 从面部色泽判断脏腑状态

问题: 如何不打开身体就知道内脏状态?
方法论的使用: "赤色小理者心小, 麤理者心大。白色小理者肺小……"面色+皮肤纹理=五脏大小。"视其外应以知其内藏则知所病矣。"
结论: 五脏的状态必然通过面色、皮肤、体态等外在特征表现出来, 存在可解读的映射关系。
结果: 建立了系统的"从外知内"诊断学, 无需侵入性检查即可评估内部状态。

案例 2: 从尺肤寒热判断疾病性质

问题: 如何快速判断疾病的寒热性质?
方法论的使用: "尺肤炬然热, 人迎大者, 当本血"——前臂皮肤灼热+颈脉盛大=血热。"掌中热者腹中热, 掌中寒者腹中寒"——手掌温度=腹部寒热。
结论: 远端(手臂)的体征可以反映核心(腹部)的状态, 存在远程映射。
结果: 通过简单的触诊即可判断内部寒热, 无需复杂仪器。

A2 — 触发场景 (Future Trigger) ★

黑箱推断: 需要了解某个不透明的系统/人/组织的内部状态, 只能通过外部信号。
间接评估: 无法直接测量目标变量, 需要找到代理指标。
模式识别: 观察到一组外部表现, 需要推断背后的原因或机制。

语言信号

"从外面怎么知道里面是什么情况"
"看不到内部, 只能从外面判断"
"这些表现说明了什么"
"如何通过XX来判断YY"

与相邻 skill 的区分

与 four-seas-regulation 的区别: 司外揣内是诊断方法(怎么看), 四海调节是调节策略(怎么调)。先用司外揣内诊断, 再用四海调节确定调哪个枢纽。

E — 可执行步骤 (Execution)

列出可观测信号
- 列出所有可直接观测的外部信号, 评估每个信号的信息量和可靠性。
- 完成标准: 至少列出3个不同类型的可观测信号。
建立信号-状态映射
- 为每个信号建立与目标内部状态的对应关系(正相关/负相关/条件相关)。
- 完成标准: 映射关系明确, 不是随意联想, 有逻辑或经验支撑。
多信号交叉验证
- 对比不同信号的推断结果, 一致则增强信心, 矛盾则需要解释矛盾。
- 完成标准: 给出综合判断, 标注信号一致性和矛盾点。
形成推断并验证
- 综合所有信号形成对内部状态的推断, 设定验证方法(如果能验证的话)。
- 完成标准: 推断有明确依据, 有验证路径。

B — 边界 (Boundary) ★

不要在以下情况使用

可以直接观测: 如果内部状态可以直接测量(如打开机箱看硬件), 不需要间接推断。
单一信号过度解读: 一个指标不能定乾坤, "参伍合参"多信号交叉才有意义。

失败模式

假阳性: 外部信号误导(如面部发红不一定都是热证), 需要多信号验证。
映射过时: 信号-状态的映射关系可能随环境变化而失效, 需要定期校准。

作者的盲点

灵枢的外内映射基于经验归纳, 现代场景需要验证映射关系是否成立。司外揣内的思维方法(从proxy推断真实值)是通用的, 但具体的映射规则需要在各自领域重新建立。

Read the full file on GitHub · 123 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

12d ago First seen · 123 lines · 74 tokens per session scan A 49bae7bb215c

Subscribe to this mod's changes

observe-infer is a skill published in the GitHub repository dhicoc/wuyun-liuqi-skills (42 stars, last pushed 27d ago), licensed MIT. It adds 74 tokens to every session and 1,691 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Research proxy-measurement 公众号

observe-infer