nexscope-ai/nexscope-ecommerce-skills / ecommerce-reverse-product-image-search-api
ecommerce.reverse-product-image-search-api
VerifiedFind visual or keyword-derived competitors from a product image or URL across supported marketplaces. Use this skill when the user asks for this exact ecommerce workflow, analysis, data lookup, or deliverable.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 6.1 KB |
| manifest.json | config | 761 B |
| references/api.json | config | 1.5 KB |
| references/api.md | markdown | 4.0 KB |
| references/credentials.md | markdown | 984 B |
| references/data-fields.md | markdown | 1.9 KB |
| references/output-schema.md | markdown | 1.4 KB |
| references/testing.md | markdown | 1.1 KB |
| references/workflow.md | markdown | 1.3 KB |
| scripts/junglescout_sales_estimates.py | script | 11.0 KB |
| scripts/nexscope_paths.py | script | 15.9 KB |
| scripts/response_io.py | script | 19.5 KB |
| scripts/step_3_5_junglescout.py | script | 5.8 KB |
| scripts/step_4_merge_rank.py | script | 13.4 KB |
SKILL.md
View on GitHub---
name: ecommerce.reverse-product-image-search-api
version: 1.0.4
category: ecommerce
description: Find visual or keyword-derived competitors from a product image or URL across supported marketplaces. Use this skill when the user asks for this exact ecommerce workflow, analysis, data lookup, or deliverable.
---
# Reverse Product Image Search Api
## Core Concepts
Find visual or keyword-derived competitors from a product image or URL across supported marketplaces.
## Core workflow
1. Identify the marketplace, entity identifiers, date range, filters, and requested output.
2. Ask only for missing inputs that materially change the result.
3. Read `references/api.json` and select the exact package-local script for the required connector, calculation, transformation, or account workflow.
4. Execute the cataloged mode exactly: run `gateway-script` and `local-script` operations with the package-local script recorded in `upstreamPath`, and follow this document for `agent-guided` operations.
5. Validate returned fields and preserve the distinction between source facts, calculations, and recommendations.
6. Save large results through the runtime and present a concise user-facing summary with the artifact path.
## Operation Catalog
| Operation | Mode | Mutation | Callable |
|---|---|---:|---:|
| `step_3_5_junglescout` | local-script | no | yes |
| `step_4_merge_rank` | local-script | no | yes |
| `junglescout_sales_estimates` | gateway-script | no | yes |
The machine-readable source of truth is `references/api.json`. Field-level payload rules remain operation-specific; do not invent identifiers, filters, credentials, or marketplace facts.
## API Invocation
Every research-tool request must use the full path recorded in `references/api.json`, beginning with `/api/v1/tools/research/`. Set `NEXSCOPE_PROXY_BASE` to the gateway origin only; never call a legacy short path such as `/amazon/**`, `/kalodata/**`, or `/aigc/**` directly.
Run only the package-local script recorded in the selected operation's `upstreamPath`. Business/tool operations may use the Nexscope proxy; authorization/account operations use their configured Agent/Login service, and third-party connectors use their documented provider endpoint. Never invoke a script from another Skill.
## Response handling
For the research HTTP response, require a numeric outer `code` equal to `0` before using `data`. Nonzero codes are platform failures; display outer `msg` without interpreting its wording as a retry instruction. Nexscope preserves upstream messages and translates Chinese to English; successful responses may have `msg: null`. HTTP 200 alone and nested business `code` / `status` do not replace the platform check. See `references/api.md` for response paths and task-specific states.
## Usage Examples
```bash
python scripts/step_3_5_junglescout.py --help
# Then run the selected local script with its documented arguments.
```
For a cataloged mutation, obtain explicit user approval and then pass both `--confirm` and `--confirm-mutation`. Never infer confirmation from an earlier read request.
## Runtime configuration
Business/tool operations require `NEXSCOPE_PROXY_BASE` plus `NEXSCOPE_API_KEY`; authorization/account operations require the relevant `NEXSCOPE_AGENT_*` or `NEXSCOPE_LOGIN_*` base because those proxy routes are not implemented. Third-party connectors require their documented provider configuration. Install optional libraries named by a script only when that operation is selected. Read `references/credentials.md` and set only the variables required by the selected script. Never pass secrets in command arguments or persist them in output artifacts.
Use only the hosts, credentials, and endpoints implemented by the package-local script and documented in `references/credentials.md`. Never substitute an undocumented legacy host or expose secret values.
## Display Rules
- Keep full JSON in the generated `nexscope/<date>/<session>/data/` artifact.
- Show the user the relevant records, assumptions, calculations, and limitations.
- Redact authorization, tokens, cookies, secrets, and passwords recursively.
- Do not treat an HTTP 200 response as success when the outer platform `code` is nonzero; nested business fields are not platform status codes.
## Important Limitations
- Do not fabricate marketplace data or silently fill required business identifiers.
- Do not retry a mutation after an ambiguous timeout; query status first.
- Treat legal, compliance, tax, and profitability outputs as decision support, not professional advice.
- Do not claim real-account, sandbox, ZIP-install, or production validation unless that evidence exists.
## References
- Read `references/workflow.md` before executing the domain procedure.
- Read `references/api.md` for the package contract and examples.
- Read `references/testing.md` before claiming a validation level.
## Privacy and Errors
Treat marketplace, account, product, advertising, order, supplier, and generated-asset data as private. Keep credentials out of payloads and artifacts. Surface HTTP, application-envelope, upstream-business, malformed-response, and unavailable-connector failures distinctly.
## Authentication
Set the `NEXSCOPE_API_KEY` environment variable. If credentials are missing or expire, visit https://www.nexscope.ai/help/skills-external-access?co-from=skillNS to top up credits.
## API Key request rate
All Nexscope API Keys belonging to the same user share one request-rate allowance for public API execution. The allowance is measured over any rolling 1-second window. Your RPS depends on your account entitlements; do not assume every user has a fixed 1 RPS. A single Skill operation may make multiple public API requests, and each request uses this allowance.
If an execution request returns HTTP 429 with outer code `13012`, the request-rate allowance was exceeded. Honor the returned `Retry-After` before considering another request. Outer code `13013` means the concurrent-request limit was exceeded, not the rate limit; honor its `Retry-After` when supplied. Inspect the response and reconcile state before retrying. Never automatically replay a write operation or a request whose outcome is unclear.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Unverified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |