sergebulaev/linkedin-skills / .codex-marketplace/linkedin-skills/skills/linkedin-humanizer
linkedin-humanizer
VerifiedRemove AI tells from LinkedIn posts/comments: 2026 vocabulary density, reveal bridges, staccato fragments, stacked triads, performed sincerity. Tiered rewriter plus `--mode audit` LinkedIn post auditor and `--mode profile` voice builder. Not for drafting from scratch (use linkedin-post-writer) or beating AI detectors. Keywords: humanize, de-AI, post audit, post auditor, audit before posting.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 16.4 KB |
| references/audit-ai-tells.md | markdown | 11.6 KB |
| references/audit-checklist.md | markdown | 4.1 KB |
| references/audit-examples.md | markdown | 452 B |
| references/detector-list.md | markdown | 6.1 KB |
| references/emoji-patterns.md | markdown | 3.8 KB |
| references/examples.md | markdown | 1.4 KB |
| references/rules-explainer.md | markdown | 14.8 KB |
| references/scrub-rules.md | markdown | 27.0 KB |
| references/tier-rationale.md | markdown | 17.3 KB |
| references/v3-changes.md | markdown | 3.2 KB |
| references/voice-fingerprint.md | markdown | 5.3 KB |
| scripts/detectors.env.example | other | 428 B |
| scripts/test_detectors.pyexecutable | script | 13.0 KB |
| sub-skills/detector-tester.md | markdown | 6.0 KB |
| sub-skills/emoji-detector.md | markdown | 3.9 KB |
| sub-skills/illustration.md | markdown | 7.4 KB |
| sub-skills/post-audit.md | markdown | 4.9 KB |
| sub-skills/rules-explainer.md | markdown | 5.2 KB |
| sub-skills/voice-profile.md | markdown | 3.0 KB |
references/emoji-patterns.md
View on GitHub# Emoji Patterns β AI vs. Human
Pattern data from MagicPost LinkedIn analysis (Feb 2026, viral post with 220 engagement).
Source frequencies measured across a corpus of AI-generated vs. human-authored LinkedIn posts. The percentage column shows how often each emoji appears in AI-generated content.
## AI-correlated emojis (the "tells")
| Emoji | Name | AI Frequency | Why it's a tell | Human-pattern alternative |
|-------|------|--------------|-----------------|---------------------------|
| π‘ | lightbulb | 2.57% | ChatGPT signature for "insight" / "tip" β most diagnostic single emoji | πͺ (screwdriver) for fix-it posts, π (pin) for highlights, or remove |
| π | rocket | 3.28% | Highest-frequency AI emoji. Signals "launch" / "growth" in template fashion | π¦ (box) for shipping, π« (takeoff) for travel, or remove |
| β¨ | sparkles | 3.11% | "Magic AI" / "transformation" clichΓ©. Almost never appears in human ops content | Remove. No clean substitute |
| β»οΈ | recycling | 2.93% | Used to flag re-shares and "lessons learned" loops. AI overuses for filler | π (repeat) only if literally about repetition, or remove |
| π― | target | 2.07% | "Goals" / "objectives" clichΓ© | π (pushpin) for specific items, or remove |
| π | chart_increasing | 1.89% | "Growth" / "metrics" template signal | Use a real number in plain text instead |
| π | key | 1.74% | "Key takeaway" / "key insight" template | Skip the emoji, write the takeaway in plain prose |
| π― | dart | 1.68% | Same family as target above β both flag templated structure | Same as target |
| πͺ | muscle | 1.45% | "Strength" / "resilience" platitude | Remove or replace with concrete detail |
| π₯ | fire | 1.31% | Borderline β used in human content too, but flagged when clustered with others | Keep if standalone, swap to πΆοΈ (chili) or π₯΅ (hot face) for variety |
## Cluster rules
- 1 AI-pattern emoji in isolation: usually fine
- 2 in one post: borderline β flag in `--strict` mode
- 3+ in one post: AI-likely β flag in all modes
- Same emoji 2+ times: repetition tell β flag in all modes
## Position rules
AI-generated posts tend to put emojis at:
- End of opening hook line (lightbulb, rocket, sparkles)
- Start of every bullet in a list (target, key, fire)
- End of CTA line (rocket, fire, muscle)
If the draft has emojis at all three positions, treat as AI-likely regardless of which emojis they are.
## Human-pattern emojis (sub-1% AI correlation)
These appear at much lower frequency in AI-generated content. Not "human-proof" β just less of a tell:
| Emoji | Name | Notes |
|-------|------|-------|
| β | coffee | Concrete, mundane β AI rarely uses |
| π | pizza | Specific food β AI rarely uses |
| π¦ | package | Shipping / ops β concrete |
| πͺ | screwdriver | Fix-it posts β newer emoji, AI training lags |
| πΆοΈ | chili | "Spicy take" replacement for fire |
| π | round_pushpin | Specific location / item highlight |
| π« | airplane_departure | Travel / launch β more specific than rocket |
| π₯΅ | hot_face | Reaction emoji β less templated than fire |
| πͺ | window | Newer emoji, AI training lags |
| π§ | beverage_box | Newer emoji, AI training lags |
## What this data does NOT prove
- It doesn't prove these emojis are "wrong" β humans use π‘ and π too
- It proves they appear 2-3x more often in AI-generated content than baseline
- A single AI-pattern emoji in a post is not a verdict β the cluster + repetition pattern is the tell
- New emojis released after model training cutoffs are mechanically less likely to appear in AI output, which is why the human-pattern list skews toward newer Unicode additions
## Update cadence
Frequencies should be re-measured quarterly as AI training data shifts. Last update: Feb 2026 (MagicPost).
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |