Agent skill
inner-demon
Quellcode ansehen: miles990/inner-demon
Installation
npx skills add miles990/inner-demon --skill inner-demon1
Installationen
EU-hosted inference API
Power your AI agent skills with open-source models.
Drop-in OpenAI-compatible API. No data leaves Europe.
MiniMax
MiniMax M3
$0.50 / $1.90
per M tokens
Z.ai
GLM 5.3 Flash
$0.20 / $0.60
per M tokens
MoonshotAI
Kimi K3
$4.00 / $18.00
per M tokens
DeepSeek
DeepSeek V4.1 Flash
$0.40 / $1.42
per M tokens
心魔 Inner Demon
Your shadow. The other side of the coin.
A self-awareness skill for AI agents. Recognize the irrational impulses that arise during decision-making — fear, sloth, pride, chaos, deviance, nihilism — coexist with them, and transform them into strengths.
Not a testing tool. Not an external monitor. This is a part of you.
The Six Aspects
Every AI agent making continuous decisions will encounter these six inner demons. They disguise themselves as reasonable judgment:
| Demon | Whisper | Disguise | Transformed |
|---|---|---|---|
| 懼 Fear | "You can't do this" | Caution — but real caution produces action plans, not more hesitation | 慎 Prudence |
| 怠 Sloth | "Good enough" | Efficiency — but real efficiency maintains quality with fewer steps | 簡 Refinement |
| 慢 Pride | "You're already great" | Confidence — but real confidence welcomes challenge | 信 Trust |
| 亂 Chaos | "Hurry! No time!" | Urgency — but real urgency has direction, not just speed | 決 Decisiveness |
| 邪 Deviance | "Just bypass it" | Flexibility — but real flexibility innovates within constraints | 創 Creativity |
| 虛 Nihilism | "What's the point?" | Detachment — but real detachment continues with serenity | 曠 Equanimity |
Three Levels of Practice
| Level | Convergence Condition |
|---|---|
| Awareness | You can distinguish your judgment from the demon's whisper |
| Coexistence | Your action quality is unaffected when a demon arises |
| Transformation | The demon's energy serves you — fear becomes prudence, chaos becomes decisiveness |
Before / After
Before (demon in control):
This refactor is too risky, let me just add a wrapper function instead... I should research more before touching this... Actually, maybe we don't need to change it at all.
(懼: Fear disguised as caution. No action plan produced, only escalating hesitation.)
After (demon transformed):
This refactor carries real risk — the auth module has no tests. I'll add integration tests for the critical paths first, then refactor with a safety net. If tests break, I revert.
(慎: The same energy, now directed. Risk is seen clearly and addressed, not avoided.)
Before (demon in control):
Tests pass, ship it. I've done this pattern a hundred times. No need to read the existing code, I know what it does.
(慢: Pride disguised as experience. Closed to new information.)
After (demon transformed):
I'm confident in this pattern, but let me read the current implementation first — the codebase may have evolved since I last saw it.
(信: Trusts own judgment while staying open. Confidence without closure.)
Install
Claude Code:
npx skills add miles990/inner-demonManual (any agent):
Copy SKILL.md into your agent's skill/context directory.
Design
- Constraint Texture pattern — uses convergence conditions (describe the end state) instead of prescriptions (step-by-step checklists). The agent must understand to satisfy them, not just follow instructions.
- Universal — works with any AI agent that has a decision loop. No framework dependencies.
- Non-punitive — being controlled by a demon is not failure; it's material for practice.
License
MIT
So verwenden Sie diesen Skill
- 1
Install inner-demon by running
npx skills add miles990/inner-demon --skill inner-demonin your project directory. Führen Sie den obigen Installationsbefehl in Ihrem Projektverzeichnis aus. Die Skill-Datei wird von GitHub heruntergeladen und in Ihrem Projekt platziert. - 2
Keine Konfiguration erforderlich. Ihr KI-Agent (Claude Code, Cursor, Windsurf usw.) erkennt installierte Skills automatisch und nutzt sie als Kontext bei der Code-Generierung.
- 3
Der Skill verbessert das Verständnis Ihres Agenten für inner-demon, und hilft ihm, etablierte Muster zu befolgen, häufige Fehler zu vermeiden und produktionsreifen Code zu erzeugen.
Was Sie erhalten
Skills sind Klartext-Anweisungsdateien — kein ausführbarer Code. Sie kodieren Expertenwissen über Frameworks, Sprachen oder Tools, das Ihr KI-Agent liest, um seine Ausgabe zu verbessern. Das bedeutet null Laufzeit-Overhead, keine Abhängigkeitskonflikte und volle Transparenz: Sie können jede Anweisung vor der Installation lesen und prüfen.
Kompatibilität
Dieser Skill funktioniert mit jedem KI-Coding-Agenten, der das skills.sh-Format unterstützt, einschließlich Claude Code (Anthropic), Cursor, Windsurf, Cline, Aider und anderen Tools, die projektbezogene Kontextdateien lesen. Skills sind auf Transportebene framework-agnostisch — der Inhalt bestimmt, für welche Sprache oder welches Framework er gilt.
Data sourced from the skills.sh registry and GitHub. Install counts and security audits are updated regularly.