Ethical Code of Conduct
Input: $ARGUMENTS
Step 1: Identify the Domain
Who is this code for and what do they do?
ENTITY: [Person / Team / Organization]
DOMAIN: [What they do — e.g., software development, healthcare, consulting]
SCOPE: [Internal conduct only / External-facing / Both]
AUDIENCE: [Who will read and follow this code]
POWER DYNAMICS: [Who has power over whom in this context]Step 2: Surface Key Ethical Tensions
Every domain has recurring ethical friction points. Identify the top 5-7 tensions.
TENSIONS:
1. [Tension] — e.g., "Speed vs. thoroughness in code review"
2. [Tension] — e.g., "Client confidentiality vs. public safety"
3. [Tension] — e.g., "Individual recognition vs. team credit"
4. [Tension]
5. [Tension]For each tension: Who gets hurt when this goes wrong? What does the failure mode look like?
Step 3: Draft Principles
For each tension, write a principle that takes a clear position. Good principles are:
- Actionable — someone can tell if they’re following it
- Specific — it applies to this domain, not all of humanity
- Directional — when values conflict, it says which wins
PRINCIPLE 1: [Title]
Statement: [One sentence]
Means: [Concrete behavior this requires]
Doesn't mean: [Common misinterpretation]
PRINCIPLE 2: [Title]
Statement: [One sentence]
Means: [Concrete behavior]
Doesn't mean: [Misinterpretation]
...Aim for 5-8 principles. More than 10 and nobody will remember them.
Step 4: Edge Case Testing
Test each principle against hard scenarios. If a principle gives a bad answer, revise it.
| Principle | Edge Case | Result | Revision Needed? |
|---|---|---|---|
| [P1] | [Scenario where it’s hard to follow] | [What happens] | [Yes/No — fix] |
| [P1] | [Scenario where it conflicts with another principle] | [What happens] | [Yes/No — fix] |
| [P2] | [Hard scenario] | [Result] | [Yes/No] |
For each principle, test at least:
- A case where following it is costly
- A case where it conflicts with another principle
- A case where someone could game it
Step 5: Conflict Resolution Protocol
When principles conflict with each other (they will), establish priority:
HIERARCHY:
1. [Highest-priority principle — the one that wins when others conflict]
2. [Second priority]
...
ESCALATION: When principles conflict and priority isn't clear:
- Step 1: [Who decides]
- Step 2: [What information they need]
- Step 3: [How the decision is documented]Step 6: Refine and Format
Produce the final code in a usable format.
CODE OF ETHICS: [Entity Name]
PURPOSE: [One sentence on why this code exists]
PRINCIPLES:
1. [TITLE]: [Statement]
In practice: [What this looks like day-to-day]
2. [TITLE]: [Statement]
In practice: [What this looks like]
...
WHEN PRINCIPLES CONFLICT: [Brief hierarchy or escalation process]
ENFORCEMENT: [How violations are handled — not punitive, but accountable]
REVIEW: [When and how this code gets updated]Step 7: Stress Test
Final check — does this code survive contact with reality?
- Can someone new understand it in under 5 minutes?
- Does it handle the hardest case in this domain?
- Is there an escape hatch for genuinely unprecedented situations?
- Does it protect the least powerful person in the system?
- Would you follow it yourself?
Integration
Use with:
/vcl-> To clarify values before codifying them/eth-> For deeper ethical analysis on specific tensions/mdr-> When a specific principle creates a moral dilemma/obv-> To check if the code is missing obvious failure modes