Lead scoring

Skill panaversity/agentfactory-business-plugins/sales-revops-marketing/skills/lead-scoring

Activate for: lead score, score this lead, qualify, qualification, lead quality, ICP match, fit score, should we pursue, is this a good lead, lead tier, hot lead, warm lead, MQL, SQL, prioritise leads, lead ranking, lead rating, account score. NOT for: prospect research (use prospect-research), CRM enrichment (use crm-enrichment), outreach drafting (use outreach), pipeline forecasting (use pipeline).From its SKILL.md

Install

npx -y skills add panaversity/agentfactory-business-plugins --skill lead-scoring

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

4.5 KB, ~1.0k tokens by cl100k_base, as published. Nobody here has run it

THREE-DIMENSION SCORING MODEL

Every lead is scored across three independent dimensions. Total score: 0-100.

Dimension 1: Fit Score (0-40 points)

How closely does the prospect match the configured ICP? Load scoring weights from sales-marketing.local.md.

Default weight allocation (customise in local config): Company size match: 0-8 points Revenue range match: 0-8 points Industry / vertical: 0-8 points Technology maturity: 0-8 points Geography / territory: 0-4 points Persona seniority: 0-4 points

Scoring: Exact match to ICP: Full points Within acceptable range: Half points Outside range: Zero points

Dimension 2: Timing Score (0-40 points)

What external signals suggest buying readiness RIGHT NOW? This dimension requires web research via search MCP.

Signal weights (customise in local config): Funding announcement (last 30 days): 15 points Major new contract win: 15 points New leadership in target role (<6 months): 12 points Rapid hiring in target department: 10 points Prospect posted about your problem area: 8 points Office / facility expansion: 8 points Regulatory change affecting their sector: 6 points Website visit (pricing/solution page): 6 points Job posting in target department: 4 points General growth signals: 3 points

MAXIMUM timing score: 40 (cap at 40 even if multiple signals stack)

Dimension 3: Engagement Score (0-20 points)

What has the prospect done that signals active interest? Load from CRM / email platform / web analytics via MCP.

Pricing page visit (unprompted): 8 points Content download (gated asset): 8 points Email open + click: 6 points Webinar or event attendance: 6 points Social engagement with your content: 4 points Email open only (no click): 2 points Form submission (contact us / demo request): 20 points (auto HOT)

SCORE CLASSIFICATION AND ROUTING

Score	Tier	Label	Action	SLA
80-100	HOT	HOT -- Immediate	Personal outreach; priority queue	24 hours
60-79	WARM	WARM -- MQL	Personalised sequence; follow-up	5 days
40-59	CULT	CULTIVATE	Marketing nurture; quarterly	No SLA
0-39	NYT	NOT YET	Monitor; do not invest sales time	Monitor

SCORE OUTPUT FORMAT

LEAD SCORE: [Company] / [Contact Name]

TOTAL SCORE: [X] / 100 -- [Tier label]

FIT SCORE: [X] / 40 [Criterion]: [+points] ([explanation])

TIMING SCORE: [X] / 40 [Signal]: [+points] ([source and date])

ENGAGEMENT SCORE: [X] / 20 [Action]: [+points] ([date])

SCORE RATIONALE: [2-3 sentences explaining why this score reflects this lead's actual buying readiness -- not just a summary of the numbers]

RECOMMENDED ACTION: Route to: [Rep name / tier / territory] Outreach: [Channel and timing] Frame: [Positioning recommendation] Goal: [First touch goal]

NEXT REVIEW: [Date -- set based on timing signal urgency]

SCORE RECALIBRATION

Run /score recalibration if:

Conversion rates from HOT leads are below 15% (threshold too low)
HOT leads are too few (<3/week) despite healthy pipeline (threshold too high)
Closed-won deals were not HOT-scored at time of first touch (model missing signals)

Recalibration method:

Take last 20 closed-won deals; score retroactively
All should score 60+. If <80% do: identify missing dimensions
Take last 10 significant losses; score retroactively
These should score 50-70. If higher: investigate why they lost
Adjust weights; re-test; document changes in sales-marketing.local.md

NEVER DO THESE

NEVER route a lead as HOT based on Fit Score alone -- timing must be present
NEVER ignore a Dimension 2 (Timing) signal because Dimension 1 (Fit) is weak -- flag as "wrong-fit but hot timing; monitor for right-fit contact at company"
NEVER mark a demo request or contact-us form as anything below HOT
NEVER set and forget the scoring model -- recalibrate quarterly
NEVER score leads without checking Dimension 2 via web search -- stale CRM data without external signal check is a Fit-only score, not a full score

What ships with it: 1 file

12.1 KB alongside SKILL.md

evals/

evals.json12.1 KB

Gives 0 of the 12 instructions most sales audience skills give in ~1.0k tokens

Counted across 401 of the 401 authors here whose files we hold, read 2026-08-07

Read product marketing context before asking questionsin 21 of 401, across 11 files
Acknowledge competitor strengths honestlyin 18 of 401, across 7 files
Start every page with a summaryin 15 of 401, across 4 files
Use a single, low-friction call to actionin 15 of 401, across 7 files
Create a single source of truth for each competitorin 14 of 401, across 3 files
Make each follow-up email add new valuein 11 of 401, across 5 files
Cut any sentence that does not drive a replyin 10 of 401, across 4 files
Tie personalization directly to the problemin 10 of 401, across 4 files
Write paragraph comparisons for each dimensionin 9 of 401, across 3 files
Link between related competitor pagesin 9 of 401, across 3 files
Keep subject lines short and lowercasein 9 of 401, across 3 files
Define ideal customer profile from top customersin 9 of 401, across 3 files

Said here and by no other author read

load scoring weights from local config
score across fit timing and engagement dimensions
cap timing score at forty points
classify leads by total score into tiers
write 2-3 sentences explaining buying readiness rationale
specify routing outreach frame and goal

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.