zal4dw/car-hunter/negotiation-coach

Claude Code and agent skill by zal4dw/car-hunter

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

This skill should be used when the user asks for help negotiating a used car purchase: "help me negotiate", "what should I offer", "is this price fair", "how do I haggle with the dealer", "draft an offer email", "the dealer said X, what do I say", "should I take this deal", or any request about the buying/offer/haggling stage for a car they found via car-hunter. Builds a data-backed negotiation strategy from the tracked market data: evidence pack, offer anchors, scripts, and counters to dealer tactics.

Install

npx vskill@latest install zal4dw/car-hunter/negotiation-coach

SKILL.md

Negotiation Coach

Help the user negotiate a specific car purchase, grounded in the market data car-hunter has collected. The numbers come from the bundled brief script; this skill turns them into a strategy, concrete scripts, and live coaching as the negotiation unfolds.

Prerequisites

  • A car profile and at least one search snapshot ({profile_name}-searches/ with dated CSVs). Without data, still coach - but say clearly the advice is general, not evidence-based, and offer to run /search-cars first.
  • The target listing's listing_id. If the user describes the car instead ("the blue RS in Leeds for £62k"), match it against the latest CSV by variant, price, and location, and confirm the match before proceeding.

Step 1: Build the evidence pack

bash
python3 "${CLAUDE_PLUGIN_ROOT}/scripts/negotiation_brief.py" \
--profile "${CLAUDE_PLUGIN_DATA}/profiles/{profile_name}.json" \
--dir "{profile_name}-searches" \
--listing {listing_id}

The JSON brief contains: the target's market position (asking vs modelled price), days on market, price-change history across snapshots, ranked negotiation levers with strengths, suggested offer anchors (opening / target / walk-away), and the closest comparables with price gaps. If the listing is missing from the latest snapshot, it may have sold - tell the user and offer to re-run the search.

Step 2: Present the assessment

Lead with the verdict, then the evidence:

  1. Is the price fair? Asking vs modelled market price, in plain terms ("£2,400 over the odds" / "already priced £1,100 under market - this one is about speed, not haggling").
  2. The levers, strongest first - relay each lever from the brief with its number attached. Evidence beats adjectives: "it has sat for 64 days" lands harder than "it seems stale".
  3. The anchors: opening offer, realistic target, walk-away. Explain the basis. Remind the user the model cannot see condition, service history, or tyres - an inspection can move every number.
  4. The comparables: the two or three closest, with price gaps and links - this is the user's printable evidence.

Step 3: Coach the negotiation

Tailor to where the user is in the process:

Before contact - advise on channel (email/phone gives a paper trail and removes showroom pressure), timing (month-end and quarter-end help; a car already reduced and 60+ days in stock helps more), and viewing strategy (view and test drive before talking numbers; note defects calmly - each is a quantifiable deduction).

Making the offer - draft the message for them on request. Structure: genuine interest, specific evidence (days listed, comparables, market position), one clear number, polite willingness to walk. Offer the total "out-the-door" price including fees, not a monthly payment.

Handling dealer responses - prepare counters for the standard moves:

  • "Another buyer is coming this afternoon" - fine; the offer stands if that falls through.
  • "The price is already reduced" - acknowledge, cite the comparables and days listed; the market disagrees.
  • "Best I can do" on price - pivot to non-price value: tank of charge/fuel, mats, new MOT, fresh service, tyre replacement, extended warranty, delivery.
  • Admin/doc fees appearing late - re-anchor on out-the-door total; the agreed number includes everything.
  • Monthly-payment framing - always pull back to total price; finance is a separate negotiation (and dealers earn commission on it - being open to their finance can win a price concession; check early-settlement terms before using that card).
  • Part-exchange muddying the water - negotiate the car's price first, the PX value second, never as one blended number.

Closing - get the agreed spec, price, and included extras in writing before paying a deposit; confirm the deposit is refundable pending inspection/HPI check; for distance sales note the 14-day return rights; never feel bad about walking - the supply lever exists because there genuinely are other cars.

Step 4: Stay in the loop

Offer to: draft the next reply when the dealer responds, re-assess if the dealer counters with a number ("is £61,250 now a good deal?" - compare against the anchors), and re-run the brief after the next search to see if the car dropped again or sold.

Important Notes

  • Be honest when the user has little leverage (fresh listing, priced under market, scarce variant) - the right play there is speed and decisiveness, not hardball that loses the car.
  • All prices in GBP; UK consumer context (Consumer Rights Act 2015 for dealer sales, distance-selling rules). Do not present legal points as legal advice.
  • The model's numbers are estimates from listing data. Condition, history, and ownership count - always recommend an inspection and HPI check before money moves.
  • Use UK English throughout.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/zal4dw/car-hunter/negotiation-coach/badge)](https://verified-skill.com/skills/zal4dw/car-hunter/negotiation-coach)

Something wrong with this skill or its page? Tell us.