The archive · Money & Fintech · Product decision · 2026
a-stock-data: zero-key A-share toolkit for AI agents hit 9.6k stars
Simon Lin packaged 19 A-share data sources into one self-contained AI-agent skill; within four months the repo passed 9.6k stars and 1.7k forks.
a-stock-data
What the business is
a-stock-data is a free Apache-2.0, self-contained SKILL.md (structured Markdown with embedded Python) that wraps 19 Chinese market-data sources — Tencent, Sina, mootdx/Tongdaxin, Eastmoney, Tonghuashun, iwencai, cninfo, baostock, Shenwan, the PBOC and NBS among them — into 54 zero-auth endpoints across 11 layers (quotes, research reports, signals, capital flows, news, fundamentals, announcements, limit-up boards, ETF options, sentiment and macro) for AI coding assistants such as Claude Code, Codex and OpenClaw.
How it started
Simon Lin — who describes himself as having financial-institution work experience plus AI product experience, building market-data tools and multi-agent systems — packaged the recurring A-share pain points named in the README (mootdx K-line parameters, Eastmoney PDF Referer headers, iwencai X-Claw authentication) into one skill file. The GitHub repo was created on 2026-05-11, and the README doubles as a job search: he states he is looking for Shenzhen AI roles in investment-research products, FDE and AI consulting, targeting large tech firms such as Tencent and financial institutions.
What happened
Development ran as a fast changelog: v2.1 added local caching for northbound-flow history; v3.0 removed the akshare wrapper and connected underlying HTTP APIs directly to cut middle-layer failure points; v3.2–v3.6 added limit-up (da ban) boards, ETF options, sentiment and investor-interaction layers, replaced anti-crawled Tonghuashun endpoints with Eastmoney equivalents, added fallback-source quick references and handled the Beijing Stock Exchange migration to 920-coded tickers; v3.7 (all endpoints re-verified on 2026-08-19) added baostock, Shenwan industry-change history, PBOC and NBS macro endpoints, locally computed chip-distribution (CYQ), valuation history back to 2016 and delisting dates. FAQ threads (issues #18, #21/#22/#29 and #30) show active users hitting real network and dependency problems, and the author says sponsors' issues get priority; no company, funding or paid tier is mentioned.
No ending yet — it is still running.
Background
a-stock-data is a free Apache-2.0, single-file AI skill by Simon Lin that packages the messy reality of Chinese A-share data — 19 sources, per-vendor auth schemes, Referer headers and IP-blocking risk — into 54 zero-auth endpoints across 11 layers, from real-time quotes and research reports to limit-up boards, ETF options, investor-interaction sentiment and PBOC/NBS macro data. Users install it by copying one SKILL.md into ~/.claude/skills/, then asking Claude Code, Codex or OpenClaw for things like a stock's valuation or a market-wide limit-up breakdown.
The bet is that packaging beats per-source integration: when every vendor API needs a key, a header or an anti-block strategy, one free, zero-auth, single-file toolkit becomes the default data layer for AI agents doing A-share research. The architecture follows that bet — direct HTTP connections instead of the akshare wrapper (removed in v3.0), sources ranked by IP-block risk with Eastmoney throttled through em_get(), and independent fallback sources for each data layer.
The repo was created on 2026-05-11 and passed 9.6k stars and 1.7k forks by the 2026-09-04 crawl, with 49 commits carrying it to v3.7.2 and all endpoints re-verified on 2026-08-19. The README is also an extended job application: Lin says he has financial-institution experience and 17k+ cumulative GitHub stars and is looking for AI investment-research, FDE or consulting roles in Shenzhen — reputation built in public is the product's current business model.
What has to be true
- Zero-auth plus one-file install removed every adoption barrier, so a Claude Code user could go from curl to a live A-share query in about two minutes.
- Ranking sources by IP-block risk and shipping fallback endpoints treated China's unstable vendor APIs as the core problem rather than an edge case.
- The author's finance background and visible evidence — all endpoints re-verified on 2026-08-19 with measured examples — made the README read like infrastructure.
- Dropping the akshare middle layer cut dependency failure points, and each changelog version answered a real breakage users reported in issues.
- The open question is whether 17k+ cumulative stars convert into the Shenzhen AI role he advertises — so far the currency earned is reputation, not revenue.
What can be applied
Give the hard default away for free: a no-key file that just works made a solo toolkit the community default; the payoff so far is reputation, not revenue.
Aftermath
As of 2026-09-05 a-stock-data is live at v3.7.2 with 9.6k stars and 1.7k forks on GitHub; its 54 endpoints across 11 layers were re-verified on 2026-08-19, the README continues in Simplified Chinese with an English translation, and the author remains an independent developer job-hunting in Shenzhen rather than a company. No funding, team or paid product is visible in the material; monetization gestures are limited to a coffee-donation line and sponsor-priority issues.
Sources
spotted an error? The archive wants to know.
Your turn
You just read one. Describe what you are building, and see who is betting on the same thing.
Free account · 3 free questions · no card