qveris-a-stock-data-layer

QVeris A-Stock Data Layer

Use this skill to preserve the A-Stock Data candidate's broad A-share data-layer intent while replacing legacy endpoints, scraping, and provider-specific dependencies with QVeris finance CAP calls plus a narrow audited Web lane for news and qualitative sentiment.

Source record:

Source Adaptation

• Preserve the original ten-layer research/data intent: market data, research reports, signals, capital/ownership, news, fundamentals, filings, limit-up/limit-down, ETF options, and investor interaction.
• Replace the original embedded Python and direct source routing with QVeris CAP calls. The source repo's endpoints, headers, anti-ban rules, mootdx, requests, stockstats, and optional iwencai key are migration context only.
• Convert original valuation flows to evidence-gated trailing or consensus-input notes; do not output target prices, upside/downside, trade setups, or option strategies.
• Convert limit-up, LHB, unlock, capital-flow, ETF-option, hot-list, and investor-Q&A layers to conditional capabilities: call them only after current cap-detail confirms a QVeris finance CAP and matching fields.
• Keep broad data-layer reports concise for users; put full call details and rejected payloads in the trace appendix.

Runtime Contract

• Use only qveris_finance.* CAP tools and QVERIS_API_KEY.
• Execute every finance data call through this Skill's scripts/qveris_finance_adapter.mjs, or through a native wrapper that runs the byte-identical adapter; never call /capabilities/query directly from the workflow.
• Default natural-language output to a Markdown user report, not a JSON object.
• Accept dry_run, max_calls, max_age, and budget_note; if omitted, default to dry_run=false, no hard max_calls limit, max_age=P1D, and a conservative budget note, then echo those controls.
• Read references/qveris-finance-data-quality-rubric.md before using QVeris payloads as evidence.
• Use references/qveris-finance-retry-policy.md for 5xx, fetch failures, 404s, payload truncation, and semantic mismatches.
• Read references/qveris-finance-capability-fallbacks.json before replacing a failed CAP. Use only a listed same-claim fallback, preserve its declared evidence downgrade, and keep unlisted capabilities missing.
• Build trace, call counts, retries, and timestamps only from saved observed_calls. Never invent an execution ID or a call that was only planned; use execution_id=null when an observed call returned no ID.
• Sanitize every output surface, including Evidence, Sources, prose, params, responses, and Trace. Strip provider names, provider API URLs, raw route/tool IDs, candidates, failover, credentials, and routing metadata recursively; the Trace row remains exactly tool_name, params, status, execution_id, fallback_used, and missing_fields.
• Sanity-check entity, market, exchange, asset type, currency, date window, fiscal period, and payload shape before using data.
• Treat a successful transport response with wrong A-share identity, wrong date window, wrong fiscal period, or too-thin bars as rejected evidence.
• Suppress target prices, upside/downside, ratings, buy/sell wording, rebalancing instructions, and trade execution plans even if present in a QVeris payload.
• Read and follow references/qveris-web-news-sentiment-policy.md. Never call qveris_finance.news_fin_tagged or qveris_finance.sentiment_text_signals; use its audited Web lane in every run mode, including benchmark and replay.