Imported from topviewai/skill (
topview-amazon-ops/SKILL.md). Install upstream withnpx skills add topviewai/skill --skill topview-amazon-ops. Copyright stays with the author.
Topview Amazon Ops
Public Amazon market / ASIN / keyword intelligence through the host's unified topview-mcp server (progressive disclosure; domain amazon).
Step 0 — Bootstrap
- Confirm the host already has MCP server
topview-mcpconnected (look foruse_amazon_data,get_tool_schema,call_tool). - If tools are missing, tell the user to add the server from repo
mcp.example.jsoninto~/.cursor/mcp.json, reload MCP — then retry. - Do not run
scripts/mcp_call.py, open SSE yourself, or use curl/raw HTTP.
Execution
Server: topview-mcp (unified). Never open SSE yourself.
Amazon data tools are deferred — they do not appear as top-level tools/list entries. Invoke via progressive disclosure:
- Read the
use_amazon_datadescription ontools/list(embedded Tools catalog). Callinguse_amazon_dataonly to list tools is optional; call it once if the catalog says it is not loaded yet (Amazon proxy warm-up). - Pick a deferred tool name from the catalog or references/tool_routing.md.
get_tool_schemawith{ "name": "<deferred>", "domain": "amazon" }.call_toolwith{ "name": "<deferred>", "domain": "amazon", "arguments": { ... } }.
Typical host tool ids: mcp__topview-mcp__get_tool_schema / mcp__topview-mcp__call_tool (Cursor may prefix user-).
Do not assume deferred names like market_research are direct top-level tools. Creative tools (topview_* / ta_*) are on the same server but belong to topview-generate — call those directly, never via call_tool.
UX Rules
- Be concise; lead with findings, not RPC noise.
- Ask for marketplace / category / ASIN / keyword when required by the tool schema.
- Never invent private Seller Central Ads/order/profit numbers.
- Match the user's language.
- After analysis, if the user wants creatives, chain to
topview-generate.
Source
Routing adapted from Marketing Studio Amazon expert skills (V032).