Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Hertwill/hw-mcp --skill skillgit clone --depth 1 https://github.com/Hertwill/hw-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/hertwill/hw-mcp/skill)<a href="https://agentmods.dev/skills/hertwill/hw-mcp/skill"><img src="https://agentmods.dev/badge/skills/hertwill/hw-mcp/skill/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/hertwill/hw-mcp/skill"><img src="https://agentmods.dev/badge/skills/hertwill/hw-mcp/skill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00123 | $0.02560 |
| Opus 5.5 | $0.00049 | $0.01024 |
| Sonnet 5.5 | $0.00025 | $0.00512 |
| Haiku 4.5 | $0.00012 | $0.00256 |
Grade A, and why
hertwill-sourcing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 211 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hertwill Sourcing Playbook
This Skill gives you the domain knowledge needed to source winning dropshipping products from the Hertwill catalog. It covers five areas: winner heuristics, EU shipping guidance, markup strategy, category-to-niche mapping with seasonal timing, and hard guardrails. Use the Hertwill MCP tools referenced throughout to execute each step.
1. Winner Heuristics
A "winner" is a product worth importing into a store. Apply these filters in order.
Margin Floor
The minimum gross margin after EU VAT and estimated ad spend is 60%.
- EU VAT ranges from 20% to 27% depending on the member state. Always use the target country's actual rate.
- Estimated ad spend (Meta/Google) typically runs 15-25% of retail price for cold-traffic dropshipping.
- Never do mental math. Use
calculate_marginwith the product's cost, your intended retail price, the target VAT rate, and your ad-spend estimate. If the result is below 60%, the product fails the margin test -- raise the retail price or skip it.
Stock State
- Never recommend
out_of_stockproducts. Check viaevaluate_productorget_product. If stock isout_of_stock, skip the product entirely. - Flag
lowstock with an explicit warning. The product may sell out before the store gains traction. Tell the user the risk. - Only
in_stockproducts should be presented as confident winners.
Variant Count
- Prefer products with 3 or more variants (size, color, style). More variants means more SKU surface on a listing, more keyword coverage, and higher average order value potential.
- Single-variant products are acceptable but note the limitation: fewer upsell angles and less listing richness.
Seasonal Fit
- Cross-reference the seasonal calendar in Section 5 before calling a product a winner. A summer product evaluated in October is a losing pick regardless of margin.
- If the product is seasonal, confirm you are within or approaching the relevant season before recommending it.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 211 lines · 123 tokens per session scan A 7f531a029779
hertwill-sourcing is a skill published in the GitHub repository Hertwill/hw-mcp (5 stars, last pushed 4mo ago), licensed MIT. It adds 123 tokens to every session and 2,560 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
hertwill-sourcing compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| hertwill-sourcingHertwill/hw-mcp | Skill | 5 | 4mo ago | MIT | A | 123 | Claude Code, Codex |
| amazon-reviews-api-skillbrowser-act/skills | Skill | 6.3k | 3d ago | MIT | A | 124 | Claude Code, Codex |
| amazon-competitor-analyzerbrowser-act/skills | Skill | 6.3k | 3d ago | MIT | A | 48 | Claude Code, Codex |
| asc-subscription-localizationrorkai/app-store-connect-cli-skills | Skill | 1.1k | today | MIT | A | 60 | Claude Code |
| external-purchasesrshankras/claude-code-apple-skills | Skill | 867 | today | MIT | A | 76 | Claude Code |
| food-orderBitterbot-AI/bitterbot-desktop | Skill | 2.5k | today | MIT | A | 53 | Claude Code, Codex |
| Product Photoshoot Workflowjau123/MeiGen-AI-Design-MCP | Skill | 1.8k | yesterday | MIT | A | 111 | Claude Code |