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 Vladimir-Human/ru-marketplace-mcp --skill taobao-connectorgit clone --depth 1 https://github.com/Vladimir-Human/ru-marketplace-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/vladimir-human/ru-marketplace-mcp/taobao-connector)<a href="https://agentmods.dev/skills/vladimir-human/ru-marketplace-mcp/taobao-connector"><img src="https://agentmods.dev/badge/skills/vladimir-human/ru-marketplace-mcp/taobao-connector/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/vladimir-human/ru-marketplace-mcp/taobao-connector"><img src="https://agentmods.dev/badge/skills/vladimir-human/ru-marketplace-mcp/taobao-connector.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00076 | $0.00580 |
| Opus 5.5 | $0.00030 | $0.00232 |
| Sonnet 5.5 | $0.00015 | $0.00116 |
| Haiku 4.5 | $0.00008 | $0.00058 |
Grade A, and why
taobao-connector 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 1mo 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- taobao-connector — 88% identical, 21 lines differ
How it starts
The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Taobao Connector
Reads Taobao inside the operator's own Chrome over CDP. The search page is a
client-side React app whose data layer is the signed mtop API — every XHR wants
a sign computed from the _m_h5_tk cookie, and unsigned probes answer
FAIL_SYS_TOKEN_EMPTY. There is no anonymous tier; the site's own JS signs
requests natively in a real browser.
When to use
- Search Taobao by Chinese or English text
- One item's price, shop, sales label, description images
Tools available
taobao_search(query, page)— listings. Prices are in YUAN (CNY), never converted: price_cny is None when hidden — never 0.taobao_card(item_id_or_url)— one item card. item ids are 9-13 digit strings; pass a bare id or an item.taobao.com URL.
Not an MCP tool: taobao_selfcheck() is a tri-state canary that renders one
live search page. It is CLI-only — marketplace-mcp doctor runs every
connector's canary at once.
Gotchas
- Prices stay in yuan. Comparing against ruble sources needs an explicit rate; a baked-in one would go silently stale.
- A detected login/CAPTCHA wall returns
challenge_requiredandrequires_user_action=true. Complete it in the scraping profile before retrying the same operation. Failed payloads are not cached; a retry reads the browser again. Successful results still use the normal TTL cache. - With
CHROME_CHALLENGE_HANDOFF_Senabled,handoff_expires_atconfirms that the challenged search/card tab is retained. Complete its interaction, then repeat the same tool arguments in the same MCP session before expiry. No new navigation is issued for that resume; expiry never extends on retries. - The operator's Chrome must be running with CDP (scripts/start_chrome_cdp.sh).
DSH activation
In DeepSeek Harness, the default profile exposes only compare_prices and
compare_sources through the cheap compare mount. Per-marketplace tools and
marketplace_sources require RU_MARKETPLACE_MCP_FULL=1 and a profile restart;
do not call them in the default mode.
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.
- 1mo ago Changed · +6 lines 2f2ead924402
- 1mo ago Changed · +6 lines 99a9b0e7ebf4
- 1mo ago First seen · 34 lines · 76 tokens per session scan A b99015a92802
taobao-connector is a skill published in the GitHub repository Vladimir-Human/ru-marketplace-mcp (163 stars, last pushed yesterday), licensed MIT. It adds 76 tokens to every session and 580 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
commerce-detail-page-copy
A writing tool for Korean e-commerce product detail pages. It can generate a full 13-section page, assess an existing page, or create two persona-based copy sets.
cs-channel-message
A channel-message writing tool for search ads, advertising, customer relationship messages, and app notifications. It uses the NCM sequence—Need, Channel, Moment, Message, and CTA—to adapt wording to where and when customers see it.
commerce-detail-page-planner
A strategy brief for deciding what a Korean e-commerce product page should say, show, and prioritise. It defines the main message, opening, page structure, sales channel approach, and photo checklist before copy, code, or images are made.
commerce-ad-claim-compliance-kr
A pre-publication check for Korean advertising and product-page wording. It reviews claims against South Korea's advertising, food-safety, and e-commerce laws.
commerce-automation-audit
A planning and consulting tool for finding automation opportunities in an online store. It reviews work such as products, pricing, orders, stock, marketing, and business data, then proposes priorities and a phased roadmap.
commerce-integrated-strategy
An e-commerce planning skill that combines customer research, product messaging, and store data into a one-page sales strategy with three priority actions.