Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Build a LinkedIn content system for thought leadership. Use when someone needs to establish authority, attract inbound leads, or build a consistent content presence. Covers positioning, content pillars, formats, and posting rhythm.
.claude/skills/brianrwagner-linkedin-authority-builder/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 76% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 89% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 59% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 133% | 0% |
Here's what most people get wrong about LinkedIn: they're trying to go viral.
Viral doesn't pay your bills. Being remembered by the right 500 people when they need what you do — that pays your bills.
This skill builds a content system around consistent positioning and clear pillars — not hacks.
Detect from context or ask: "Starter plan, full content system, or 90-day build?"
| Mode | What you get | Best for | |------|-------------|----------| | quick | 3 content pillars + 1-week posting starter plan | Getting unstuck, first week of consistency | | standard | Full content system: pillars, formats, posting rhythm + first week written | Building a repeatable presence | | deep | Full system + 90-day calendar + 10 posts written + engagement playbook | Serious authority-building campaign |
Default: standard — use quick if they just need to start posting. Use deep if they're committing to LinkedIn as a growth channel.
Before generating any strategy, load:
positioning-basics output if available)Sequencing gate: If positioning isn't clear yet, stop and ask: > "Before building a content strategy, I need your one-liner: 'I help specific audience] with specific outcome] through unique approach].' Do you have this locked, or should we nail positioning first with positioning-basics?"
Do not build a content system for an unclear position — the content will be unfocused.
Also suggest: > "Run linkedin-profile-optimizer if you haven't — the content we build needs a profile that converts the traffic."
Before recommending any content, reason through:
Output a brief strategy assessment: > "You're a role] targeting audience] with X hours/week] and a goal of outcome]. I'll build a 3/5 post/week] strategy anchored to X] pillars. The main gap in your current approach: specific gap]."
Each pillar must pass all 4 tests:
Pillar ratio:
Output format per pillar:
Pillar: [Name]
Ratio: [%]
Content types: [frameworks / stories / case studies]
Example hook: "[First line of a real post]"
Connection to goal: [how this drives the stated outcome]| Format | Best For | Engagement Level | |---|---|---| | Story | Connection, memorability | High | | Framework/List | Authority, credibility | High | | Hot take | Reach, visibility | Variable | | Case study/proof | Credibility, late-stage trust | Medium | | Behind-the-scenes | Relatability, trust | Medium |
Recommended weekly mix:
Story Post:
[Hook — the moment or realization]
[Setup — quick context]
[Tension — what was hard or went wrong]
[Turn — the insight]
[Lesson — the takeaway]
[Question — drives engagement]Framework Post:
[Hook — bold claim or problem statement]
[Why this matters — 1-2 sentences]
[The X-step framework:]
1. [Step + brief explanation]
2. [Step + brief explanation]
3. [Step + brief explanation]
[Key insight or summary]
[CTA or discussion question]Hot Take:
[Controversial statement]
[Your reasoning — 2-3 sentences]
[The nuance people miss]
[What to do instead]
[Question to drive comments]Generate a 4-week calendar with actual dates (not generic day names):
| Date | Pillar | Format | Hook (first line) | Status | |---|---|---|---|---| | YYYY-MM-DD] | Pillar] | Framework | "First line]" | Draft | | YYYY-MM-DD] | Pillar] | Story | "First line]" | Draft |
Fill 4 full weeks. Generic "Week 1, Monday" output is not sufficient.
Also output 5 starter post hooks ready to write immediately — these break the blank-page problem:
1. [Hook]
2. [Hook]
3. [Hook]
4. [Hook]
5. [Hook]After generating the strategy, evaluate:
Flag issues: "Pillar 3 ('general business tips') doesn't connect to your stated goal of attracting SaaS founders. Replace with something more specific."
After 30 days, review with these questions:
Adjust based on data:
markdown## LinkedIn Strategy: [Name] — [Date] ### Positioning Alignment [One-liner + assessment of clarity] ### Content Pillars [3-5 pillars with ratio, examples, connection to goal] ### Weekly Rhythm - Posts/week: [X] - Best times: [e.g., Tue/Thu 8am EST] - Active commenting: [X min/day] ### Format Mix [Breakdown with rationale] ### 4-Week Content Calendar [Table with real dates + hooks] ### 5 Starter Posts (Write These First) [Hooks with format labels] ### Engagement Plan [Who to engage with, how much time, what to say] ### Self-Critique Notes [Issues flagged + recommended fixes] ### 30-Day Review Triggers [What to measure and when to adjust] ### Cross-References - linkedin-profile-optimizer (run before publishing) - content-idea-generator (for ongoing idea generation) - voice-extractor (to ensure posts sound authentic)
Skill by Brian Wagner | AI Marketing Architect | brianrwagner.com
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 32,406 | 26,902 | -17% | 1 | 1 | 0% | 5,060 | 5,996 | +18% | 0 | 0 | — |
case-07 | fail→fail | 15,984 | 8,676 | -46% | 1 | 1 | 0% | 2,501 | 3,221 | +29% | 0 | 0 | — |
case-08 | fail→pass | 18,606 | 21,608 | +16% | 1 | 1 | 0% | 2,966 | 5,214 | +76% | 0 | 0 | — |
case-09 | fail→pass | 14,905 | 18,191 | +22% | 1 | 1 | 0% | 2,478 | 4,686 | +89% | 0 | 0 | — |
case-10 | pass→pass | 18,863 | 17,093 | -9% | 1 | 1 | 0% | 2,914 | 4,556 | +56% | 0 | 0 | — |
case-02 | fail→fail | 22,610 | 23,425 | +4% | 1 | 1 | 0% | 3,572 | 5,583 | +56% | 0 | 0 | — |
case-03 | fail→pass | 35,591 | 28,031 | -21% | 1 | 1 | 0% | 5,621 | 6,600 | +17% | 0 | 0 | — |
case-04 | pass→fail | 10,257 | 10,481 | +2% | 1 | 1 | 0% | 1,613 | 3,485 | +116% | 0 | 0 | — |
case-05 | pass→pass | 18,484 | 13,604 | -26% | 1 | 1 | 0% | 2,936 | 3,922 | +34% | 0 | 0 | — |
case-06 | pass→pass | 14,420 | 17,509 | +21% | 1 | 1 | 0% | 2,201 | 4,387 | +99% | 0 | 0 | — |
case-22 | pass→pass | 15,450 | 11,462 | -26% | 1 | 1 | 0% | 2,138 | 3,589 | +68% | 0 | 0 | — |
case-20 | fail→pass | 26,561 | 28,257 | +6% | 1 | 1 | 0% | 3,937 | 6,270 | +59% | 0 | 0 | — |
case-11 | pass→fail | 29,028 | 29,297 | +1% | 1 | 1 | 0% | 4,543 | 6,525 | +44% | 0 | 0 | — |
case-12 | pass→fail | 11,834 | 5,240 | -56% | 1 | 1 | 0% | 1,859 | 2,796 | +50% | 0 | 0 | — |
case-13 | fail→pass | 8,987 | 5,981 | -33% | 1 | 1 | 0% | 1,226 | 2,862 | +133% | 0 | 0 | — |
case-14 | pass→pass | 21,165 | 24,958 | +18% | 1 | 1 | 0% | 3,213 | 5,714 | +78% | 0 | 0 | — |
case-21 | pass→pass | 14,021 | 8,426 | -40% | 1 | 1 | 0% | 1,969 | 3,190 | +62% | 0 | 0 | — |
case-15 | fail→fail | 12,395 | 14,578 | +18% | 1 | 1 | 0% | 1,799 | 4,031 | +124% | 0 | 0 | — |
case-16 | fail→fail | 13,552 | 11,543 | -15% | 1 | 1 | 0% | 2,032 | 3,525 | +73% | 0 | 0 | — |
case-17 | fail→pass | 22,889 | 33,594 | +47% | 1 | 1 | 0% | 3,422 | 6,231 | +82% | 0 | 0 | — |
case-18 | pass→pass | 17,782 | 28,162 | +58% | 1 | 1 | 0% | 2,636 | 5,952 | +126% | 0 | 0 | — |
case-19 | pass→pass | 21,465 | 14,678 | -32% | 1 | 1 | 0% | 3,057 | 4,053 | +33% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.