/crest

Anti-Patterns

Purpose: Self-branding anti-pattern detection rules and platform-specific pitfalls. Read when: REFINE phase of any mode — apply AP checks to all outputs before delivery.

Core Anti-Patterns (AP-1 ~ AP-7)

Applied to every Crest output. All must pass before delivery.

AP-1: Resume Dump

Signal: Raw skill/experience lists without narrative, context, or impact framing.

Fix:

• Convert each item to impact format: "Achieved [quantitative result] in [outcome] with [Technology]"
• Limit to 5-8 core technologies aligned with niche
• Add narrative arc: why you chose this path, what you learned

Severity: Medium — fixable by restructuring existing content.

AP-2: Vanity Metrics

Signal: Metrics cited for impressiveness without demonstrating real impact.

Fix:

• Pair every vanity metric with an impact metric (see metrics-guide.md)
• Replace "stars" with "weekly downloads" or "contributor count"
• Replace "followers" with "engagement rate" or "inbound opportunities"

Severity: Medium — misleads audience and recruiters.

AP-3: Niche Absence

Signal: Positioning so broad that it provides no differentiation.

Fix:

• Apply Tech×Domain×Perspective framework (positioning-frameworks.md)
• Target L2 minimum, L3 ideal
• If genuinely multi-skilled, pick one niche for branding and mention others as secondary

Severity: High — without niche, all other branding efforts are diluted.

AP-4: Channel Scatter

Signal: Inconsistent messaging, positioning, or persona across platforms.

Fix:

• Create unified positioning statement and adapt (not change) per platform
• Audit all channels quarterly for consistency
• Deactivate or redirect channels you don't maintain

Severity: High — confuses audience and weakens brand recognition.

AP-5: AI Ghost

Signal: Content that sounds generated, lacks personality, or is indistinguishable from AI output.

Fix:

• Add 1+ personal anecdote per article (failure, surprise, learning moment)
• Include opinions and hot takes (with reasoning)
• Use conversational language specific to your personality
• Break templates occasionally with unique formats

Severity: High (growing) — audiences increasingly detect and distrust AI-generated content.

AP-6: Employer Leak

Signal: Public content that exposes employer confidential information.

Fix:

• Generalize: "あるFinTech企業で" instead of naming the employer
• Abstract: describe patterns and principles, not specific implementations
• Get approval: when in doubt, run content by your manager
• Time-delay: wait 6-12 months before discussing past projects in detail

Severity: Critical — can result in termination or legal action.

AP-7: Stagnation Mask

Signal: Brand that only showcases past achievements with no evidence of current growth.

Fix:

• Add "Currently exploring" section to profiles
• Share learning journey content (TIL posts, Zenn Scraps)
• Update profiles quarterly
• Mix authority content (what you know) with growth content (what you're learning)

Severity: Medium — erodes trust over time as audience perceives stagnation.

Platform-Specific Anti-Patterns

Qiita

Zenn

note

X (Twitter)

YouTube

TikTok / Instagram Reels

Anti-Pattern Audit Template

Output this with every Crest deliverable:

## Anti-Pattern Check Results

| # | Check | Status | Note |
|---|-------|--------|------|
| AP-1 | Resume Dump | ✅/⚠️/❌ | [detail] |
| AP-2 | Vanity Metrics | ✅/⚠️/❌ | [detail] |
| AP-3 | Niche Absence | ✅/⚠️/❌ | [detail] |
| AP-4 | Channel Scatter | ✅/⚠️/❌ | [detail] |
| AP-5 | AI Ghost | ✅/⚠️/❌ | [detail] |
| AP-6 | Employer Leak | ✅/⚠️/❌ | [detail] |
| AP-7 | Stagnation Mask | ✅/⚠️/❌ | [detail] |

Platform-specific checks: [list applicable platform checks]

Delivery rule: All AP-1~AP-7 must be ✅ or ⚠️ (with documented mitigation). Any ❌ blocks delivery.