agentleFS
28672032962867203296/hermes-agent/skills/hermes/thinking-visualization/SKILL.md

thinking-visualization

Skill0 starsChanged 3 months agoMITView on GitHub

思维链可视化:Agent 推理过程展开为可读的思维树。借鉴 Claude Code thinking 过程 + OpenAI o-series 思维链

The file

Files
SKILL.md

Thinking Visualization — 思维链可视化

用户最常问的:"你为什么选 A 不选 B?" 本技能让推理过程可见。

一、三层思维展开

L1:简要推理(默认,每次回复自带)

"TTS 不工作了 → 上次修过 voice.py → 先检查那里"

一句话说清推理路径。用户不需要喊"show thinking"就能看到。

L2:思维树(用户要求时展开)

TTS 不工作了
├─ 方案A: 重装 edge-tts (30%把握)
│   └─ 风险:可能不是包的问题,浪费时间
├─ 方案B: 检查 voice.py (85%把握) ✓
│   ├─ 依据:上次同样问题修过 voice.py:835
│   └─ 预期:0.5s重开麦被注释了
└─ 方案C: 换 openai-tts (10%把握)
    └─ 风险:需要API key,用户没有
→ 选择B,先验证再决定是否A

L3:完整推理链(调试时使用)

{
  "problem": "TTS不工作",
  "observations": [
    "用户说播报没声音",
    "上次类似问题在voice.py:835"
  ],
  "hypotheses": [
    {"h": "重开麦逻辑被注释", "confidence": 0.85, "evidence": "session_search"},
    {"h": "edge-tts包损坏", "confidence": 0.30, "evidence": "推测"}
  ],
  "eliminated": [
    {"h": "换TTS引擎", "reason": "用户明确用edge-tts"}
  ],
  "decision": "检查voice.py",
  "outcome": "已修复"
}

二、关键决策标注

当 Agent 做出影响较大的决策时,自动在回复中标注:

[决策] 选择修复voice.py而非重装edge-tts
[原因] 上次同样问题修过同文件,置信度85%
[替代] 如果修复无效,第二步考虑重装

三、不确定性表达

把握表达
> 90%"确定是..."
70-90%"很可能是..."
50-70%"我的判断是...但建议验证"
< 50%"据我所知...可能不准,建议查一下"

四、与 agent-observability 联动

  • thinking-visualization:事前呈现推理过程
  • agent-observability:事后回放决策链

两者互补覆盖"为什么选 A"的完整答案。

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.