Skip to content

nexscope-ai/nexscope-ecommerce-skills / ecommerce-mercado-market-intelligence

ecommerce.mercado-market-intelligence

Verified

Query Mercado Libre public market intelligence through the Nexscope research proxy. Use when the user requests this public marketplace or patent data; do not use for seller-console or store-authorized operations.

FileKindSize
SKILL.mdskill md5.5 KB
manifest.jsonconfig420 B
references/api.mdmarkdown15.6 KB
references/testing.mdmarkdown2.0 KB
scripts/damai_mercado_market_intelligence.pyscript14.9 KB
scripts/upload_image.pyscript9.6 KB

references/testing.md

View on GitHub
# Testing ecommerce.mercado-market-intelligence

- Compile `scripts/upload_image.py` and verify it uses `/api/skill-asset/presign` → presigned HTTPS `PUT` → `/api/skill-asset/confirm` without sending `NEXSCOPE_API_KEY` to the upload host.

## Static and mock gates

- Compile `scripts/damai_mercado_market_intelligence.py` with Python.
- Verify the request URL begins with `/api/v1/tools/research/`, authentication uses `Bearer`, and timeout remains compatible with the source contract.
- Mock a successful Nexscope envelope (`code=0`, business object in `data`), an outer failure, HTTP 401, HTTP 402, malformed JSON, and a network failure.
- Verify full responses are stored below `nexscope/<date>/<session>/data` and secrets are never written.

## English invocation checks

1. Query the documented public entity with the smallest valid request and summarize the result.
2. Explain the expected credit cost before requesting a second page, region, or paid operation.
3. Reject an undocumented parameter without making a request.

## Live test evidence

Record the timestamp, test base URL, redacted request, HTTP status, outer `code`, inner status, `traceId`, server-reported `X-Cost-Token` and `X-Cost-Credit`, and saved response path. A paid live call consumes credits, requires explicit approval, and must not be retried automatically.

## NEX-198 validation evidence

- Static/frontmatter/manifest and offline Mock checks: passed.
- 2026-08-27 test Curl: HTTP 200, outer `code=0`, inner `errcode=200`; raw header and body saved under `transfer/test-results/public-ecommerce-migration/`.
- External Curl did not expose upstream billing headers; backend monitoring `response.headers` supplied array-valued evidence. `X-Cost-Token=0`, calculated credits `0`, reported `X-Cost-Credit=n/a`, trace `41c7b5ca891d6804746c929f1a65017d`. Do not substitute body `costToken`.
- ZIP/local install: passed in the isolated `transfer/test-results/public-ecommerce-migration/local-install/skills` root.
- Production publication: not performed.
add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks.Stars: 115kUpdated Claude CodeCodexVerifiedInstall