How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

marketplace-advisor - Skill

Name: marketplace-advisor
Author: borghei

Imported from borghei/claude-skills (vertical-advisors/marketplace-advisor/SKILL.md). Install upstream with npx skills add borghei/claude-skills --skill marketplace-advisor. Copyright stays with the author (MIT + Commons Clause).

Marketplace Advisor

Strategic frameworks for two-sided and multi-sided marketplace founders, operators, and product leaders. Marketplaces have distinctive economics — most ecommerce / SaaS playbooks don't translate.

Keywords

marketplace, two-sided market, three-sided market, multi-sided market, supply, demand, liquidity, take rate, network effects, chicken and egg, GMV, repeat rate, fill rate, supply density

Clarify First

Before scoring, settle the inputs below. When one is missing or vague and would change the result, ask about it; when it wouldn't, use a sensible default and say which one you used.

Measurement unit: whole marketplace or one liquid segment (city/category)? Liquidity beats GMV; a blended global number can hide a broken core (sets what the score actually describes).
Category + take rate: benchmarks vary widely (Etsy ~6.5% vs Uber ~25-30%), so the take-rate-sustainability dimension is meaningless without the category.
Supply, demand, and fill rate: drive the liquidity and balance dimensions, usually the binding constraint.
Repeat rate: drives repeat-strength; a low repeat often signals off-platform leakage rather than a healthy marketplace.

Ask about the two or three inputs that matter most for this request, not the whole list. If the user wants a result straight away, produce it and state your assumptions at the top.

Quick Start

python scripts/marketplace_health_scorer.py metrics.json

Scores marketplace health across liquidity, supply/demand balance, take rate sustainability, repeat rate, and network-effect strength.

Core Workflows

Workflow 1: Marketplace Health Diagnostic

Capture key metrics: GMV, take rate, supply, demand, fill rate, repeat rate
Run: python scripts/marketplace_health_scorer.py metrics.json
Identify which dimension is the constraint (usually one of: not enough supply, not enough demand, low repeat, weak liquidity, take rate too high or too low)
Plan one focused intervention; don't try to fix all five at once

Time Estimate: 2-4 weeks per diagnostic.

Workflow 2: Chicken-and-Egg Strategy

Read references/marketplace_dynamics.md
Identify your "constrained side" (usually supply for new marketplaces)
Pick a strategy: subsidize the constrained side, single-player mode, vertical wedge, geographic concentration
Measure liquidity in the smallest viable unit (one city, one category, one buyer segment)

Time Estimate: 6-12 months for first liquid segment.

Workflow 3: Take-Rate Decision

Read references/take_rate_design.md
Benchmark category norms (eBay 10-13%, Airbnb 14%, Uber 25-30%, Etsy 6.5%, vertical B2B varies)
Decide: high take rate (full-stack with lots of value-add) vs. low take rate (commodity matching)
Plan trajectory: most marketplaces raise take rate over time as value-adds compound

Time Estimate: 4-8 weeks for first take-rate decision.

Tools

marketplace_health_scorer.py

Scores marketplace health on five dimensions: liquidity, balance, take-rate sustainability, repeat-rate strength, and supply density.

python scripts/marketplace_health_scorer.py metrics.json
python scripts/marketplace_health_scorer.py metrics.json --json

Reference Guides

references/marketplace_dynamics.md — Chicken-and-egg, liquidity, network effects, vertical / horizontal trade-offs
references/take_rate_design.md — Take rate benchmarks, when to raise / lower, full-stack vs lean

Templates

assets/marketplace_metrics_template.json — Input file for the health scorer with example values

Best Practices

Liquidity over GMV. A marketplace with $10M GMV across 1,000 cities is mostly broken; the same GMV in 5 cities can be liquid and growing.
Pick a wedge. New marketplaces almost always start narrow (single city, single category, single segment) and expand.
Measure repeat early. Marketplaces without repeat are often serving as infrastructure for off-platform transactions (a leakage problem to fix or accept).
Take rate ratchets up, rarely down. Start low to attract supply; add value-adds (payments, fulfillment, insurance, financing) that justify higher take rate.
Single-side first when possible. Some categories let you build a tool for one side that becomes a marketplace once liquid (e.g., OpenTable started as restaurant software).
Beware of weak network effects. Many "marketplaces" lack real network effects — supply on one platform doesn't make demand stickier on that platform. Examine carefully.

Scope & Limitations

In Scope:

Marketplace health scoring on liquidity, supply and demand balance, take-rate sustainability, repeat-rate strength and supply density with marketplace_health_scorer.py
Chicken-and-egg strategy: picking the constrained side, subsidies, single-player mode, vertical wedge and geographic concentration
Take-rate decisions using category benchmarks and full-stack versus lean models
Network-effect and leakage checks for two-sided and multi-sided marketplaces

Out of Scope:

General pricing design outside the take rate (see business-growth/pricing-strategy/)
Building supply or demand recruitment funnels (see marketing/landing-page-generator/)
Unit economics for a single-seller online store (see vertical-advisors/ecommerce-advisor/)
Payments, trust and safety, or marketplace platform engineering

Important Caveats:

The scorer uses only the metrics you enter for one market unit, such as a city or category. A blended global input can hide a broken segment.
Score thresholds are heuristics, and take-rate benchmarks are approximate figures for named companies. Both vary by category and over time.
A health score shows where the constraint likely sits. It does not forecast GMV or prove that network effects exist.

Integration Points

Pairs with business-growth/pricing-strategy — take rate is essentially marketplace pricing
Pairs with agents/c-level/cs-fundraising-advisor.md (agent) — marketplace investor expectations differ from SaaS
Pairs with marketing/landing-page-generator for supply / demand recruitment funnels

Install marketplace-advisor and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

{ "ocm": "1", "id": "borghei-claude-skills-marketplace-advisor", "kind": "skill", "name": "marketplace-advisor", "description": "Strategic advisory for two-sided marketplace founders on chicken-and-egg, take rates, liquidity, and network effects. Use when scoping a marketplace idea or scoring health, or mentioning marketplace, take rate, liquidity, or network effects.", "publisher": "borghei", "version": "1.0.0", "capabilities": { "domains": [ "observability", "coding" ], "tags": [ "skill-md", "marketplace", "network-effects", "take-rate", "liquidity", "two-sided", "skills-sh" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "Strategic advisory for two-sided marketplace founders on chicken-and-egg, take rates, liquidity, and network effects. Use when scoping a marketplace idea or scoring health, or mentioning marketplace, take rate, liquidity, or network effects." ], "primary": false, "metadata": { "source": { "provider": "skills.sh", "repository": "https://github.com/borghei/claude-skills", "path": "vertical-advisors/marketplace-advisor/SKILL.md", "ref": "HEAD", "url": "https://github.com/borghei/claude-skills/blob/HEAD/vertical-advisors/marketplace-advisor/SKILL.md", "key": "borghei/claude-skills/vertical-advisors/marketplace-advisor/SKILL.md" }, "license": "MIT + Commons Clause" }, "instructions": "# Marketplace Advisor\n\nStrategic frameworks for two-sided and multi-sided marketplace founders, operators, and product leaders. Marketplaces have distinctive economics — most ecommerce / SaaS playbooks don't translate.\n\n---\n\n## Keywords\n\nmarketplace, two-sided market, three-sided market, multi-sided market, supply, demand, liquidity, take rate, network effects, chicken and egg, GMV, repeat rate, fill rate, supply density\n\n---\n\n## Clarify First\n\nBefore scoring, settle the inputs below. When one is missing or vague and would change the result, ask about it; when it wouldn't, use a sensible defau", "cost": { "context_tokens": 1564 } }

marketplace-advisor

About