Purpose
Turn a loosely described partner opportunity into a disciplined go/no-go evaluation — scored across commercial, product, regulatory, and strategic fit — so BD teams stop chasing low-fit deals and concentrate effort on partnerships that move the business. Surfaces the deal-killers and diligence gaps before a single intro call is wasted.
When to Use
Invoke this skill when the user asks to:
- Screen a potential crypto exchange, wallet, protocol, or fintech partner before pursuing
- Decide whether an inbound partnership request is worth pursuing
- Compare a partner opportunity against strategic priorities and pass/proceed
- Build a go/no-go memo for a partnership pipeline review
- Identify regulatory or counterparty red flags in a proposed integration or co-marketing deal
- Prioritize between several competing partnership opportunities
Inputs Required
Before starting, collect:
- Partner name and profile: Who they are, what they do, size/stage (TVL, user count, funding, token if any)
- Opportunity type: Integration, distribution/co-marketing, liquidity, custody/infrastructure, token listing, treasury, or referral
- Proposed deal structure: What each side gives and gets (revenue share, fees, equity/token, exclusivity, etc.)
- Our strategic priorities: The 2–4 goals this partnership would serve (user growth, liquidity, geographic expansion, product gap, credibility)
- Our jurisdiction(s) and licenses: Where we operate and what we're licensed/registered for
- Known concerns or context: Anything already flagged — counterparty risk, prior conversations, competitive overlap
If jurisdiction or deal structure is missing, evaluate on best-available information and flag the gap explicitly as a diligence question rather than guessing.
Instructions
Step 1: Classify the Opportunity
Tag the opportunity type and the primary value thesis in one sentence ("This deal exists to ___"). Note whether it is inbound or outbound, and whether it is reciprocal (both sides benefit symmetrically) or asymmetric (one side carries more value/risk). State the implied time horizon: quick-win (<90 days) vs. strategic build (6+ months). This frames every downstream score.
Step 2: Score the Four Fit Dimensions
Score each dimension 1–5 (1 = strong no, 5 = strong yes) with a one-line justification. Use these rubrics:
Commercial Fit (weight 30%)
- Is there a credible, quantifiable upside (revenue, users, liquidity, fee volume)?
- Is the value exchange balanced, or are we subsidizing them?
- Are there cost/effort sinks — long integration, support burden, exclusivity that caps future deals?
- Counterparty solvency and runway: can they actually deliver and survive?
Product Fit (weight 25%)
- Does the integration fill a real product gap or just add surface area?
- Technical feasibility — API maturity, chain/standard compatibility, audit status of any smart contracts
- Does it improve or degrade our core UX (custody friction, KYC handoff, settlement latency)?
- Maintenance load: one-time build vs. ongoing dependency on their roadmap
Regulatory Fit (weight 30%)
- Jurisdictional overlap: do they operate where we're licensed, and where we're not?
- Activity classification: does the partnership pull us into money transmission, custody, securities, or VASP obligations we don't hold?
- Counterparty compliance posture: AML/KYC program, sanctions screening, prior enforcement actions, sanctioned jurisdictions exposure
- Token/asset risk: is any involved asset plausibly a security, a privacy coin, or a sanctioned/high-risk asset?
Strategic Fit (weight 20%)
- Does it directly serve one of the stated strategic priorities?
- Brand/credibility alignment — does association help or hurt our reputation?
- Competitive positioning: does it lock out a rival, or hand a competitor leverage over us?
- Optionality: does it open doors or create exclusivity/lock-in that closes them?
Step 3: Compute the Weighted Fit Score
Calculate: (Commercial × 0.30) + (Product × 0.25) + (Regulatory × 0.30) + (Strategic × 0.20), on a 1–5 scale. Show the math. Apply this band:
- 4.0–5.0 → Proceed
- 3.0–3.9 → Proceed with conditions (list the conditions)
- 2.0–2.9 → Hold / renegotiate (state what must change)
- Below 2.0 → Pass
Override rule: Any Regulatory score of 1 caps the recommendation at "Pass" regardless of the weighted total — regulatory deal-killers are not averaged away.
Step 4: Build the Red-Flag Inventory
List every material risk found, tagged by severity:
- 🔴 Deal-killer: Must be resolved or the deal is off (e.g., sanctioned-jurisdiction exposure, unlicensed activity, insolvent counterparty)
- 🟡 Mitigable: Real risk that can be contained with terms or diligence (e.g., exclusivity clause, immature API, concentration risk)
- ⚪ Monitor: Low-severity items to track but not block on
If no red flags exist in a category, say so explicitly — silence is not clearance.
Step 5: Diligence Questions
Produce 6–10 specific, answerable questions to resolve before signing, grouped by owner (Commercial / Product / Legal & Compliance). Each question should target a real unknown surfaced above — not generic ("Tell me about your business"). Frame for direct answers (e.g., "Which jurisdictions do you actively serve, and do you geo-block US persons?").
Step 6: Recommendation and Next Step
State the go/no-go in one sentence, then the single recommended next action with an owner and timeframe (e.g., "Proceed — Legal to confirm VASP registration within 2 weeks before scheduling technical scoping"). Be decisive; do not hedge.
Output Format
- Structure: Header block (partner, opportunity type, value thesis) → four-dimension scorecard → weighted score with math → red-flag inventory → diligence questions → recommendation
- Length: 700–1,000 words
- Format: Structured document — scorecard as a table, red flags and questions as labeled lists, recommendation as a short decisive paragraph
- Sections: Opportunity Classification → Fit Scorecard (4 dimensions) → Weighted Score & Band → Red-Flag Inventory → Diligence Questions → Recommendation & Next Step
- Tone: BD analyst memo — analytical, risk-aware, decisive; written for a partnerships lead who needs to act, not admire
Quality Checklist
Before delivering output, verify:
- [ ] All four dimensions are scored 1–5, each with a one-line justification
- [ ] The weighted score arithmetic is shown and matches the rubric band selected
- [ ] The regulatory override rule was applied (a Regulatory score of 1 forces "Pass")
- [ ] Every red flag carries a severity tag (🔴 / 🟡 / ⚪), and each category is explicitly cleared or flagged
- [ ] Diligence questions are specific and grouped by owner — not generic
- [ ] The recommendation is a single clear verdict with a named next action, owner, and timeframe
Notes
- Regulatory fit is weighted equal to commercial fit by design — in crypto, the deal-killer is almost always compliance, not economics.
- When inputs are thin, the highest-value output is often the diligence question set, not a confident score. Flag low-confidence dimensions explicitly.
- For comparing multiple opportunities, run each through the screener and rank by weighted score, but read the red-flag inventories side by side — a 4.1 with a mitigable exclusivity clause may beat a 4.3 with a deal-killer pending resolution.
- Counterparty solvency is frequently the overlooked dimension — a great strategic fit with a partner running out of runway is a sunk-cost trap.
How to use this skill
This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.
Installing in Claude Code:
~/.claude/skills/<folder-name>/SKILL.md— auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)<your-project>/.claude/skills/<folder-name>/SKILL.md— auto-loads only when working in that project (recommended for project-specific skills)
Restart Claude Code or open a new session to register.
Using in Claude.ai, ChatGPT, Gemini, or other AI tools:
- Drag this file into a new chat, or copy and paste its contents
- Send your task or request as your next message — the AI will apply the skill where it fits
For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills