agentleFS
DeepXivDeepXiv/deepxiv_sdk/skills/deepxiv-fa-stocks/SKILL.md

deepxiv-fa-stocks

Skill843 starsChanged 8 months agoMITView on GitHub

Point-in-time US equity data for 8,000+ tickers: daily prices vs sector and SPY, per-ticker news and event clusters, SEC filings, XBRL financials, macro series and the daily macro digest, all replayable with as_of. Use for questions about a US stock, a market theme, or US macro data.

  • Installs packages

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Always pass -d stocks: automatic domain picking is still early.

    line 89
  2. Distinctive:

    read 的 id:代码(AAPL)、macro:<series>(如 macro:CPIAUCSL)、digest:market、digest:macro,或任一条记录的 id(news:… / evt_… / filing:…)

    line 93
  3. Distinctive:

    search:文字按相关性(filters.source = all \| news \| events \| filings),source=stocks 按主题找股票;query 为 '*' 或全大写代码时按时间倒序列出(insiders 只能这样)

    line 94

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

DeepXiv · 美股数据 (stocks)

NASDAQ point-in-time US equity data. US equities: prices, news, SEC filings, insider trades, financials, macro series and events, all with as_of point-in-time replay (only data visible at that moment).

Setup

pip install -U deepxiv-sdk      # deepxiv registers a token on first use

With a registered key (https://data.rag.ac.cn/register), search and read are fair use: 60/min, 5 at once, 5,000/day; on the auto-registered token they spend the deepxiv daily limit at the credits listed below. ask draws on the daily ask allowance (resets 00:00 UTC; a site-wide daily total can run out early at peak times), or 3 lifetime trial asks on the auto-registered token. On 429, back off for Retry-After seconds. Data goes to stdout; cost, quota and sources go to stderr. Exit 0 ok, 1 API error, 2 usage error. The general CLI guide is the deepxiv-cli skill.

Start here

deepxiv fa read stocks NVDA   # price vs sector ETF and SPY, recent events, news, filings
deepxiv fa read stocks NVDA -e news   # the ticker's news feed
deepxiv fa read stocks NVDA -e financials   # XBRL financials, as filed
deepxiv fa read stocks NVDA -p as_of=2026-09-04   # the same view as of a past day
deepxiv fa search stocks "AI data center power" -F source=stocks   # which tickers a theme touches
deepxiv fa search stocks "chip export restrictions" -k 5   # news + events + filings, ranked together
deepxiv fa search stocks "earnings guidance" -F ticker=TSLA -F source=events
deepxiv fa read stocks digest:macro   # today's macro releases, rates, FX, commodities
deepxiv fa read stocks macro:CPIAUCSL   # a macro series with vintages and nowcast
deepxiv fa ask "CoreWeave 最近一周为什么大跌?结合价格、新闻和财报说明" -d stocks   # cited analysis

Free samples: ids AAPL, NVDA, macro:CPIAUCSL; queries chip export restrictions, cloud server.

Search

deepxiv fa search stocks "query" [-F FILTER]... [-k N] [--offset N] [--mode M] [--head] [--json]

Modes: keyword. Up to 100 hits. Cost 2 (with --head: 3). An empty query is rejected; use "*" to list by filters only. Filters combine with AND; an undeclared field exits 2.

filtertype-F syntaxmeaning
sourcekeywordk=vWhat to search: all (default; news + events + filings ranked together, and the first page also returns stocks: the top 5 matching tickers) | news | events | filings | stocks (find stocks by theme / business) | insiders (time-ordered only) Values: all, news, events, filings, stocks, insiders.
tickerkeywordk=vTicker symbol
as_ofkeywordk=vPoint in time (ISO date or datetime), passed through unchanged
datedatetimek>=n, k=n..mTime range {gte, lte} → upstream from / to

Advanced filters (see deepxiv fa spec stocks): news_kind, form, event_type, semantic, with_returns, side, min_value.

Read

deepxiv fa read stocks ID [ID...] [-l brief|detail|full] [-e EXTRA] [-p k=v]... [-f json|md|html]
levelcreditsreturns
brief1Ticker overview: latest price, recent events, news, filings
detail (default)1Ticker overview: latest price, recent events, news, filings
full2Ticker digest input (every item carries a citable id; built on demand, ~45 s)
-e extracreditsreturns
prices1Prices: interval=1d | 1m (IEX minute bars)
financials1Financials (XBRL, as filed; later restatements hidden)
events1Events (merged from news / 8-K / Form 4)
news1News
filings1SEC filings
insiders1Insider open-market trades (Form 4 P/S)
features2Features (price / event derived)

At -l full, -p section=NAME reads one section and -p characters=N the first N characters.

Facets

deepxiv fa facets stocks FIELD lists the legal values and counts, free. Use it instead of guessing a spelling. Fields: source, news_kind, form, event_type, semantic, side.

Ask

deepxiv fa ask "question" -d stocks [-e low|medium|high] [-p k=v]... [--no-stream] [--json]

Ask about US equities: companies by theme, a ticker's prices and news flow, events, insider trades, financials, macro; questions about a past moment are answered as_of that moment. Citations are upstream record ids ([stocks:<id> title]); answers end with data sources and a not-investment-advice note

Always pass -d stocks: automatic domain picking is still early.

What this corpus will and will not tell you

  • read 的 id:代码(AAPL)、macro:<series>(如 macro:CPIAUCSL)、digest:market、digest:macro,或任一条记录的 id(news:… / evt_… / filing:…)
  • search:文字按相关性(filters.source = all | news | events | filings),source=stocks 按主题找股票;query 为 '*' 或全大写代码时按时间倒序列出(insiders 只能这样)
  • as_of:search 放 filters.as_of,read 用查询参数 as_of;不带 = 现在。结果缓存 60 秒
  • 每条记录带 license(取值和说明见下面的 license · 各行)
  • as_of 口径 · prices: daily bar of D visible after D 16:00 ET; 1m IEX bar visible at ts+1min
  • as_of 口径 · news: visible_at (live: first_seen_at; backfilled: crawl_at + 336 s)
  • as_of 口径 · filings: live 8-K / Form 4: acceptance (or first_seen_at if later); date-only filings: 22:00 ET of the date
  • as_of 口径 · financials: XBRL facts with filed date <= as_of (filed date counts from 22:00 ET)
  • as_of 口径 · macro: ALFRED vintage at as_of; daily FRED series: value dated X visible from X+1 17:00 ET; Treasury curve from 18:00 ET (backfill) / first seen (live); monthly releases after 16:00 ET on release day for the daily cut
  • as_of 口径 · events: cluster/member first_seen_at; partial clusters re-derive direction and summary from visible members
  • as_of 口径 · returns: only horizons whose end session closed before as_of
  • as_of 口径 · features: row D uses only data visible at D 16:00 ET; rows computed after as_of are recomputed
  • as_of 口径 · digest: as_of capped at the end of the session day; each section follows its source rule
  • license · sec_public: SEC EDGAR, public
  • license · us_gov_public: US federal agencies / Federal Reserve Board, public
  • license · cleveland_fed_public: Cleveland Fed nowcast, public
  • license · tiingo_licensed: Tiingo news/prices, licensed for redistribution (was tiingo_internal before 2026-10-19)
  • license · fed_bank: regional Reserve Bank / FRED-computed series (terms of use)
  • license · third_party: copyrighted series redistributed by FRED (ICE BofA, Cboe, Moody's, UMich, S&P, NAR, Freddie Mac)
  • license · wire_rss_summary: press-release RSS summaries
  • 覆盖缺口 · news history starts 2026-04-19
  • 覆盖缺口 · Form 4 before 2026-10-16 arrives with the fadata history (2026-10-23)
  • 覆盖缺口 · IEX 1-minute bars from 2026-10-19, focus list only
  • 覆盖缺口 · fundamentals: SEC XBRL filers only (no ADRs / funds)
  • 覆盖缺口 · analyst signals come from news stance (no broker feed, no consensus)
  • macro 序列(read macro:<series>):{'id': 'A191RL1Q225SBEA', 'title': 'Real Gross Domestic Product', 'frequency': 'Q', 'license': 'us_gov_public'}, {'id': 'AAA', 'title': "Moody's Seasoned Aaa Corporate Bond Yield", 'frequency': 'M', 'license': 'third_party'}, {'id': 'AWHAETP', 'title': 'Average Weekly Hours of All Employees, Total Private', 'frequency': 'M', 'license': 'us_gov_public'}, {'id': 'BAA', 'title': "Moody's Seasoned Baa Corporate Bond Yield", 'frequency': 'M', 'license': 'third_party'}, {'id': 'BAMLC0A0CM', 'title': 'ICE BofA US Corporate Index Option-Adjusted Spread', 'frequency': 'D',
  • 上游 1.0-beta(2026-10-19):license tiingo_internal renamed tiingo_licensed: Tiingo data is now licensed for redistribution; statement timeout (12 s) returns 504 with Retry-After: 60 instead of 500; GET /v1/search and GET /v1/stocks/search results cached for 60 s per parameter set; /v1/meta: base and routes[].free_samples are host-relative paths (no internal address, no doubled /v1)

Citing

Cite records as 上游每条记录的 id(px: / evt_ / news: / filing: / form4: / fin: / macro:…@vintage)原样引用,并注明 license. Say plainly when the corpus has nothing.


Generated by scripts/gen_fa_skills.py from deepxiv fa spec stocks --json on 2026-10-19. Field names and prices can change; deepxiv fa spec stocks is the live source.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.