Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Schelling Fence

ASecurity

You face a continuum from acceptable to unacceptable:

22 stars
0 votes
0 copies
0 views
Added 9/20/2026
content-marketinggo

Works with

cli

Security Analysis

A100/100

Scanned 9/20/2026

Install to Claude Code

$npx -y skills add lev-os/agents --skill schelling-fence --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Schelling Fence?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Schelling Fence
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/lev-os-schelling-fence/badge)](https://www.skillsdirectory.com/skills/lev-os-schelling-fence)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: schelling-fence
description: You face a continuum from acceptable to unacceptable:
---

# Schelling Fence

## Overview

**Strategic precommitment to defend arbitrary boundaries on slippery slopes**: Establish a conspicuous, defensible line that takes on significance as a coordination point, preventing gradual erosion toward undesirable outcomes.

**Core insight**: When facing slippery slopes, the strongest defense isn't always at the most justifiable position—it's at the most obvious, unambiguous line where everyone can coordinate to hold the boundary.

**Source**: Scott Alexander (Slate Star Codex), 2012, building on Thomas Schelling's concept of focal points

## The Framework

### The Slippery Slope Problem

You face a continuum from acceptable to unacceptable:
- No restrictions → reasonable restrictions → excessive restrictions → totalitarian control
- Never lie → white lies → significant lies → pathological lying
- Staying up reasonable hours → slightly late → very late → all-nighter → ruined sleep schedule

**Challenge**: Each marginal step seems small and justifiable in isolation, but the cumulative effect is disastrous.

### The Schelling Point Solution

**Schelling point**: A salient, obvious boundary that serves as a natural coordination point even without communication.

**Schelling fence**: A deliberately established Schelling point used as a precommitment device to resist incremental erosion.

**Key properties**:
1. **Conspicuous**: Easy to recognize, hard to mistake
2. **Binary**: Clear which side of line you're on
3. **Symbolic value**: Takes on significance beyond inherent properties
4. **Coordination**: Multiple parties/your future self can align on defending it
5. **Credible commitment**: Failing to hold this line means all positions up to next Schelling point are vulnerable

### Why Arbitrary Lines Work

**Not defending the optimal position**—defending the most **defensible** position:

- "No alcohol before 5pm" is more defensible than "moderate drinking only" (what's moderate?)
- "Zero tolerance for lying in engineering specs" is more defensible than "don't exaggerate too much"
- "Bedtime at 11pm always" is more defensible than "get enough sleep"

The first example in each pair is more arbitrary but clearer—that clarity makes it holdable.

## When to Use

**Personal commitment devices**:
- Habit formation (gym 5x/week, not "regularly")
- Avoiding temptation (no social media on phone, not "use less")
- Sleep/work boundaries (hard stop at 6pm, not "reasonable hours")

**Interpersonal boundaries**:
- Professional ethics (never accept gifts from vendors, not "small gifts okay")
- Relationship limits (certain topics off-limits, not "be respectful")
- Management decisions (all expenses require approval, not "use good judgment")

**Policy and governance**:
- Constitutional rights (absolute free speech, not "reasonable restrictions")
- Safety regulations (zero tolerance, not "minimize incidents")
- Institutional norms (mandatory recusal with conflicts, not "be aware of bias")

**Slippery slope prevention**:
- Whenever incremental movement toward bad outcome is easy
- When future you will face temptation to rationalize "just this once"
- When multiple parties need to coordinate on a boundary
- When erosion happens gradually and imperceptibly

## Implementation Steps

### 1. Identify the Slippery Slope

What's the undesirable end state you're trying to avoid?
- Complete sleep deprivation
- Totally compromised professional ethics
- Feature bloat that kills product simplicity
- Technical debt that makes codebase unmaintainable

Map the continuum from current state to disaster state.

### 2. Find Natural Schelling Points

Look for conspicuous boundaries along the continuum:

**Good Schelling points**:
- Round numbers (bedtime at 11pm, not 10:47pm)
- Categorical distinctions (weekdays vs. weekends)
- Bright lines (zero instances, not "very rare")
- Existing norms/traditions (industry standards)
- Simple rules (always/never, not "usually")

**Bad Schelling points**:
- Vague thresholds ("reasonable" "moderate" "appropriate")
- Subjective judgment calls (case-by-case basis)
- Complex conditions (if X and Y but not Z...)
- Invisible boundaries (hard to verify compliance)

### 3. Choose Conservative Fence Location

**Critical insight**: Place the fence **before** the truly problematic region.

**Why**: If you try to defend right at the edge of disaster, small violations are catastrophic. If you defend early, small violations are just small violations.

Example:
- **Too close**: "No drinking and driving" (violation = immediate danger)
- **Better**: "No drinking within 8 hours of driving" (violation = buffer remains)
- **Schelling fence**: "Never drink on nights before early meetings" (simple, defensive)

Build in margin for error.

### 4. Establish Symbolic Significance

Make the boundary **mean something** beyond its literal position:

**Explicit precommitment**: "I'm setting this boundary and I will defend it"

**Public commitment**: Tell others—social cost of violation reinforces boundary

**Link to values**: "This represents my commitment to X principle"

**Create ritual**: Make crossing the boundary require deliberate ceremony/justification

The boundary's power comes from its significance, not just its position.

### 5. Defend Vigorously

**Zero tolerance for violations**: Every exception weakens the fence.

**Resist "just this once"**: If it's justified this time, it'll be justified next time.

**Escalate defense**: Treat boundary violations as more serious than the marginal harm—you're defending the fence itself, not just the immediate issue.

**Make exceptions explicit**: If you genuinely need to move the fence, do it deliberately and establish a new one—don't just let it erode.

### 6. Repair Breaches Immediately

When the fence is violated:

1. **Acknowledge clearly**: Don't pretend it didn't happen
2. **Re-establish boundary**: Return to fence position, don't stay at violation point
3. **Strengthen if needed**: Add accountability or make boundary more conspicuous
4. **Learn**: What pressure led to breach? Reinforce those defenses

Every violation you tolerate becomes precedent for the next.

## Common Pitfalls

**Defending indefensible positions**: Putting the fence at an awkward, hard-to-explain place. Pick obvious boundaries even if they're not theoretically optimal.

**Too many fences**: If you have 20 bright-line rules, none are bright. Choose the critical boundaries.

**Justifying exceptions**: "This situation is special" is how fences erode. Exceptions should be extremely rare and involve moving the fence deliberately, not ignoring it.

**Arguing the fence position**: Getting into debates about whether the specific line is optimal. The fence's value is in its defense, not its specific location—don't undermine it by admitting it's arbitrary.

**Not making boundaries visible**: A fence you can't see or verify compliance with doesn't work. Make violations obvious.

**Gradual movement**: Slowly shifting the fence 5 minutes at a time. Defeats the purpose—fence should stay put or move in deliberate, discrete jumps.

## Real-World Applications

**Product development**: "No features that require >2 clicks" prevents gradual UX complexity creep. Each individual extra click seems fine, but the fence holds simplicity.

**Engineering standards**: "All code requires tests" prevents gradual erosion to "most code has tests" to "important code has tests" to "tests are optional."

**Personal productivity**: "No phone use before 9am" prevents checking "just one thing" spiral. Arbitrary but defensible.

**Free speech policy**: "No content moderation based on viewpoint" is clearer than "moderate extreme content"—the fence prevents incremental expansion of "extreme."

**Commitment devices**: "Zero alcohol during work week" easier to maintain than "moderate drinking"—no decisions, just a rule.

**Behavior change**: "Gym every weekday morning" beats "exercise regularly"—the fence is conspicuous, violations are obvious.

## Power Moves

**Pre-establish fences before temptation**: Set bedtime policy when well-rested, not when tired at midnight. Set expense rules when solvent, not when vendor offers "special deal."

**Use existing Schelling points**: Don't create arbitrary boundaries from scratch—anchor to existing round numbers, natural categories, social norms. Leverages existing salience.

**Make fence public**: Social pressure reinforces personal commitment. "I don't drink on school nights" is easier to maintain when friends know the rule.

**Celebrate fence defense**: Each time you successfully hold the boundary in face of temptation, reinforces its significance. "I wanted to stay up, but I held the 11pm line."

**Create friction at fence**: Add verification, reporting, or ceremony required to cross. Makes violations deliberate, not accidental.

**Multiple parties share fence**: Team agreement on boundary makes defense easier—"we don't ship without tests" is stronger than individual commitment.

## Advanced Technique: Fence Maintenance

**When to move a fence**:
- New information shows fence is in wrong place (too conservative or too permissive)
- Context change makes old fence irrelevant
- Successfully holding fence for long period allows moving closer to optimal

**How to move a fence without eroding it**:
1. Make movement explicit, deliberate decision
2. Establish new fence at specific new position
3. Communicate clearly: "Old rule was X, new rule is Y as of [date]"
4. Defend new fence as vigorously as old one
5. Don't move incrementally—jump to new stable position

**Never**: Just ignore the fence when inconvenient and hope no one notices. That destroys the fence's credibility.

## Related Frameworks

- **Commitment Devices**: Schelling fences are a type of commitment device using salience
- **Sorites Paradox**: The philosophical problem Schelling fences solve (heap of sand: when does removing grains stop being a heap?)
- **Precommitment Strategies**: Ulysses tying himself to mast—limiting future options
- **Trigger-Action Plans**: "When X, then Y" rules similar to fence-defense protocols
- **Bright-Line Rules**: Legal/ethical boundaries that are unambiguous
- **Chesterton's Fence**: Before removing a boundary, understand why it was erected

## Why "Fence" Metaphor Works

**Physical fence properties**:
- Visible: You know when you're crossing it
- Binary: You're on one side or the other
- Defended: Crossing requires effort/intention
- Symbolic: Represents property boundary beyond just physical barrier

**Schelling fence inherits these**: The rule isn't just a guideline—it's a defended boundary with significance beyond its specific position.

## Sources

- [LessWrong - Schelling Fences on Slippery Slopes](https://www.lesswrong.com/posts/Kbm6QnJv9dgWsPHQP/schelling-fences-on-slippery-slopes) (Scott Alexander, 2012)
- [Beeminder Blog - Schelling Fences](https://blog.beeminder.com/schelling/)
- [Thomas Schelling - The Strategy of Conflict](https://en.wikipedia.org/wiki/The_Strategy_of_Conflict) (original focal point concept)

Attribution

lev-oslev-os
View sourceMore from lev-os →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Postiz

Postiz is a tool to schedule social media and chat posts to 28+ channels X, LinkedIn, LinkedIn Page, Reddit, Instagram, Facebook Page, Threads, YouTube, Google My Business, TikTok, Pinterest, Dribbble, Discord, Slack, Kick, Twitch, Mastodon, Bluesky, Lemmy, Farcaster, Telegram, Nostr, VK, Medium, Dev.to, Hashnode, WordPress, ListMonk

21281 votes

Serp Analysis

SERP analysis techniques for intent classification, feature identification, and competitive intelligence. Use when analyzing search results for content strategy.

2831 votes

On Page Seo Auditor

This skill performs detailed on-page SEO audits to identify issues and optimization opportunities. It analyzes all on-page elements that affect search rankings and provides actionable recommendations.

1821 votes

Brand

Brand voice, visual identity, messaging frameworks, asset management, brand consistency. Activate for branded content, tone of voice, marketing assets, brand compliance, style guides.

1240080 votes

Release Announcement

Write a release announcement — changelog, blog post, in-app note, or social post — that leads with user impact, names the audience, and includes upgrade/migration steps without filler.

805540 votes
View all in content-marketing →