Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when the user wants B2B lead generation and sales pipeline support: prospecting, outbound lists, ICP account lists, cold email, lead scoring, MQL/SQL handoff, sales enablement, pitch decks, objection handling, demo scripts, CRM pipeline, or RevOps. Trigger on 'find leads', 'prospect list', 'cold email', 'outbound', 'sales deck', 'lead scoring', 'CRM', 'MQL', 'SQL', or 'pipeline'.
.claude/skills/minhnv0807-33-b2b-lead-gen-global/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 187% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 25% | 0% |
This skill connects marketing to sales: define ICP, find accounts and contacts, score leads, write outreach, create sales assets, and set MQL/SQL handoff. The goal is qualified pipeline, not a large email list.
Read .agents/product-marketing-context-global.md if available. Ask up to 4 questions: ICP, deal size/ACV, target market/region, and current outbound channels. Use CRM exports or existing lists as baseline when available. For deeper guidance, read references/b2b-lead-gen-playbook.md.
| User needs | Mode | |------------|------| | "Find prospects" | Prospecting | | "Write outreach emails" | Cold outreach | | "Score leads" | Lead scoring | | "Handoff to sales" | RevOps handoff | | "Build a sales deck / objections" | Sales enablement |
Clarify:
| Source | Best for | |--------|----------| | LinkedIn/Sales Nav | B2B role/title targeting | | Apollo/Clay/ZoomInfo | Account and contact data | | Google Maps/local directories | Local SMB | | GitHub/job posts | Devtool/SaaS intent | | G2/Capterra/review sites | Competitor/category demand | | Website visitors/forms | Warm intent |
Use 100 points:
| Group | Points | |-------|--------| | ICP fit | 35 | | Intent trigger | 25 | | Pain/proof | 15 | | Contact quality | 15 | | Timing | 10 |
Only push to outreach when score is at least 60/100.
Default 4-touch sequence:
Avoid spam, exaggerated claims, and fake personalization.
For high-potential accounts, create:
Every lead should have:
markdown# B2B Lead Gen Plan — [Brand] ## 1. ICP | Segment | Fit criteria | Pain | Exclusion | ## 2. Lead sources | Source | Query/filter | Data needed | Owner | ## 3. Scoring model | Signal | Points | Why it matters | ## 4. Prospect table | Account | Contact | Trigger | Score | Suggested opener | Next step | ## 5. Outreach sequence ### Email 1 Subject: Body: ### Follow-ups ## 6. Sales enablement assets | Asset | Persona | Purpose | Draft notes | ## 7. CRM handoff | Stage | Entry criteria | Owner | SLA |
09-customer-insight-global: persona, pain, JTBD.31-offer-design-global: B2B/high-ticket offer.14-email-marketing-global: nurture after capture.08-competitor-research-global: competitor triggers and alternative positioning.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 47,301 | 32,481 | -31% | 1 | 1 | 0% | 7,142 | 5,584 | -22% | 0 | 0 | — |
case-02 | fail→pass | 53,145 | 34,912 | -34% | 1 | 1 | 0% | 8,194 | 6,202 | -24% | 0 | 0 | — |
case-03 | fail→pass | 55,141 | 31,947 | -42% | 1 | 1 | 0% | 8,259 | 6,284 | -24% | 0 | 0 | — |
case-04 | fail→fail | 17,372 | 19,306 | +11% | 1 | 1 | 0% | 2,952 | 4,101 | +39% | 0 | 0 | — |
case-05 | fail→pass | 10,724 | 13,145 | +23% | 1 | 1 | 0% | 816 | 2,340 | +187% | 0 | 0 | — |
case-06 | pass→pass | 20,295 | 22,674 | +12% | 1 | 1 | 0% | 3,289 | 4,639 | +41% | 0 | 0 | — |
case-07 | pass→pass | 20,112 | 21,761 | +8% | 1 | 1 | 0% | 3,132 | 3,610 | +15% | 0 | 0 | — |
case-08 | pass→pass | 17,150 | 19,210 | +12% | 1 | 1 | 0% | 2,746 | 3,202 | +17% | 0 | 0 | — |
case-09 | fail→pass | 19,561 | 17,931 | -8% | 1 | 1 | 0% | 2,914 | 3,655 | +25% | 0 | 0 | — |
case-10 | fail→fail | 16,591 | 14,513 | -13% | 1 | 1 | 0% | 2,690 | 3,355 | +25% | 0 | 0 | — |
case-11 | pass→pass | 16,104 | 13,826 | -14% | 1 | 1 | 0% | 2,478 | 3,136 | +27% | 0 | 0 | — |
case-21 | fail→fail | 15,201 | 14,883 | -2% | 1 | 1 | 0% | 2,610 | 3,452 | +32% | 0 | 0 | — |
case-12 | pass→pass | 11,802 | 9,444 | -20% | 1 | 1 | 0% | 2,014 | 2,489 | +24% | 0 | 0 | — |
case-13 | fail→pass | 18,085 | 16,315 | -10% | 1 | 1 | 0% | 2,895 | 3,702 | +28% | 0 | 0 | — |
case-14 | fail→fail | 11,421 | 2,811 | -75% | 1 | 1 | 0% | 1,680 | 1,407 | -16% | 0 | 0 | — |
case-15 | pass→pass | 8,793 | 9,456 | +8% | 1 | 1 | 0% | 1,310 | 2,351 | +79% | 0 | 0 | — |
case-22 | fail→fail | 19,216 | 20,248 | +5% | 1 | 1 | 0% | 3,347 | 4,396 | +31% | 0 | 0 | — |
case-16 | pass→pass | 17,230 | 17,307 | +0% | 1 | 1 | 0% | 2,635 | 3,663 | +39% | 0 | 0 | — |
case-17 | fail→pass | 12,910 | 8,823 | -32% | 1 | 1 | 0% | 2,111 | 2,513 | +19% | 0 | 0 | — |
case-18 | fail→fail | 13,594 | 11,509 | -15% | 1 | 1 | 0% | 2,251 | 2,892 | +28% | 0 | 0 | — |
case-19 | fail→pass | 11,116 | 8,264 | -26% | 1 | 1 | 0% | 1,901 | 2,342 | +23% | 0 | 0 | — |
case-20 | fail→fail | 21,817 | 15,577 | -29% | 1 | 1 | 0% | 3,651 | 3,598 | -1% | 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 +36 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.