Deal Desk
Deal Desk reviews individual inbound deals before close: it scores margin and risk, routes discount approval to the correct named human approver, and flags contract landmines in redlined terms. It never auto-approves; every output is a numeric scorecard plus a routing recommendation to a named human.
✨ What it does
- Scores deals 0-100 across margin, risk, strategic value, commercial fit, and term shape, assigning APPROVE/REVIEW/ESCALA
- Routes discount percentages to a named approver chain (AE → Manager → Director → VP → CFO/CRO) with estimated cycle days
- Detects 10 contract landmines (uncapped indemnity, MFN, perpetual license-back, missing DPA, NET-60+, etc.) with severit
- Supports industry profiles: saas, enterprise-software, services, marketplace.
- Allows policy threshold overrides via input JSON.
- Provides a forcing-question library for disciplined deal review.
🎯 When to use it
- Sales requests a discount exceeding AE authority and needs approval routing.
- A customer redlines the MSA and you need triage before sending to legal.
- You need a defensible margin breakdown for CFO sign-off on a deal.
- An RFP response involves multi-year terms and you need to score the deal shape.
- A renewal expansion bundles a discount and you must verify policy fit.
🚀 How to use
Trigger: /cs:grill-commercial
Trigger by asking to review a specific deal, e.g., 'Review this discount request' or 'Triage this redlined MSA'. Provide deal details (ARR, term, discount, payment terms, customer tier, strategic flags) and optionally a structured JSON file. Run the three scripts in sequence: deal_scorer.py, discount_approval_router.py, terms_redliner.py. Example prompts:
Review this deal: 28% discount on $500k ARR, enterprise-software profile, customer redlined uncapped indemnity.
Run deal-desk on my_deal.json with profile saas.
📄 Output: A deal-desk review packet combining a margin/risk scorecard, a named approver chain with cycle days, and a ranked list of contract landmine findings with counter-language.
📦 Add this skill to Claude Code
# 1. Get the skills repo
git clone --depth 1 https://github.com/alirezarezvani/claude-skills /tmp/claude-skills
# 2. Copy this skill into your project (or ~/.claude/skills for all projects)
mkdir -p .claude/skills
cp -r /tmp/claude-skills/commercial/skills/deal-desk .claude/skills/deal-deskSkill source: commercial/skills/deal-desk/SKILL.md
⚠️ Good to know
Requires an existing commercial policy (discount bands, payment norms, indemnity caps) and structured terms JSON; it is not a substitute for full legal review or for calculating the optimal discount.
❓ FAQ
Does Deal Desk ever auto-approve a deal?
No. Every verdict, including APPROVE, names the human approver(s) who must sign; the output is a recommendation only.
What happens if a deal has a high score but contains an uncapped indemnity clause?
Critical signals like uncapped indemnity override the composite score, forcing a DECLINE regardless of margin.
Can I use Deal Desk to calculate the right discount for a deal?
No. It routes a discount already proposed by the AE or customer; pricing logic is handled by the pricing-strategist skill.
🤖 Overview, features, install steps and FAQ were generated from the project's SKILL.md on Sep 4, 2026. Always check the original source before running commands.