Vibe Research is a local financial research workspace in which an AI agent gathers market data, performs multi-step analysis, and preserves reports, evidence, calculations, and research history. It is for investment research across Chinese, US, and Hong Kong stocks, including market reviews, company studies, portfolios, debates, and backtesting. The catalogue contains skills and an instruction for working with this research agent.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/simonlin1212/vibe-research/data-accessnpx skills add simonlin1212/Vibe-Research --skill data-accessgit clone --depth 1 https://github.com/simonlin1212/Vibe-ResearchWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/simonlin1212/vibe-research/data-access)<a href="https://agentmods.dev/skills/simonlin1212/vibe-research/data-access"><img src="https://agentmods.dev/badge/skills/simonlin1212/vibe-research/data-access.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00127 | $0.04105 |
| Opus 5 | $0.00063 | $0.02053 |
| Sonnet 5 | $0.00025 | $0.00821 |
| Haiku 4.5 | $0.00013 | $0.00411 |
Grade A, and why
data-access scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
A 股零鉴权取数(data-access)
本 skill 是 AGENTS.md §0 第 1 条("禁止凭记忆生成数据")与 §5("取数只用登记脚本")的落地:每个数字都来自本次运行的脚本调用,原始响应落盘,证据带齐契约字段。脚本只取数、不做算术(单季拆分 / TTM / 同比 / 分位一律交给 calc/)。
1. 调用方式
python3 .agents/skills/data-access/scripts/<script>.py --symbol 300308 --out-dir .local/runs/<run-id>
(示例代码 300308 为 Phase 0 验收标的,仅作命令行示例,不代表任何推荐。)
--symbol接受300308/SZ300308/300308.SZ(前后缀二选一,矛盾即报错,绝不猜市场);输出统一 6 位 + 市场SH|SZ|BJ。--out-dir给运行目录:原始响应自动写到<out-dir>/raw/,结构化结果写到<out-dir>/fetch/<script>.json;不给则只打印 JSON 到 stdout(仍计算 sha256)。- 退出码:
0ok(主源成功)/2partial(走了备源或部分字段缺失,看extra.degraded与errors)/3failed(关键数据全部失败)。非 0 不是"没数据可以编",是"如实记缺口"。 - 依赖:
scripts/requirements.txt(requests / pandas / lxml / akshare / baostock);腾讯、深交所、东财 K 线只用标准库。需要联网;在 Codex 沙箱内运行时由编排器开启网络权限。 - 东财系请求全部经
common.em_get:跨进程串行(文件锁覆盖整个请求生命周期,任一时刻最多一个东财请求在途,间隔 ≥1s + 抖动)、403 不重试、代理失败自动直连重试、push2断连自动轮询push2delay。编排器并行启动多个脚本也会在锁上排队。
2. 脚本登记表
| 脚本 | 拿什么 | 主源 | 备源 | 研究中的地位 |
|---|---|---|---|---|
fetch_quote.py |
现价 / 昨收 / 涨跌幅 / 换手 / PE_TTM / PE 静 / PB / 流通市值 / 总市值 / 成交额;僵尸报价疑似 is_stale(命中 → partial 且估值类 evidence 带 note;停牌 / 废码 / 盘前三者之一,可用性由 SOP 结合交易日历判定) |
腾讯 qt.gtimg.cn | 东财 push2(delay)(同样做僵尸判定,period 取源端行情时间) | ★ 必需(估值分子) |
fetch_profile.py |
名称 / 上市日 / 在市状态 / 证监会行业 / 东财行业 / 总股本 / 流通股 / 市值 | 腾讯 + baostock | 东财 push2(delay)(可选增强,失败不拖垮) | ★ 必需 |
fetch_financials.py |
最近 N 报告期累计值:营业总收入 / 归母净利润 / 扣非净利润 / 基本 EPS;关键字段 × 最近 8 期完整性校验 | 新浪财务摘要(akshare) | 新浪利润表直连(无扣非;主源部分缺失时只补营收 / 归母,补齐也算走备源 → partial) | ★ 必需(扣非×4 PE / TTM 同比) |
fetch_estimates.py |
一致预期 EPS(FY T / T+1 / T+2):均值 / min / max / 机构数 | 同花顺 worth.html | 东财研报逐篇预测(非一致预期,partial) | ★ 必需(前瞻 CAGR) |
fetch_pe_history.py |
PE_TTM / PB 日频序列(默认 5 年)→ raw CSV;最新值 | baostock | —(北交所不支持) | ○ 可选(TTM PE 分位) |
fetch_announcements.py |
最近 N 条公告标题 / 日期 / PDF 链接 | 深市:深交所官方;沪市 / 北交所:东财 | 深市备源东财 | ○ 可选(风险 / 反证线索) |
fetch_kline.py |
日 K 前复权序列 → raw;最新收盘(逐行校验,坏行剔除→partial) | 腾讯 fqkline | 东财 push2his(本机网络常断) | ○ 可选(stale 二次验证时用) |
fetch_trade_calendar.py |
全市场交易日历:last_trading_day / previous_trading_day / is_today_trading_day / session_phase(pre_open·trading·post_close·non_trading_day)/ reference_quote_day(此刻新鲜报价应有的日期)(evidence symbol=MARKET, market=CN) | baostock query_trade_dates | — | ★ 必需(判定报价日期差异是休市 / 盘前还是个股停牌) |
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- scripts/common.py 15 KB runs code
- scripts/core/__init__.py 0 B runs code
- scripts/core/cdp.py 27 KB runs code
- scripts/core/exa_client.py 8.3 KB runs code
- scripts/core/retrieval.py 24 KB runs code
- scripts/core/stdio_utf8.py 3.3 KB runs code
- scripts/fetch_announcements.py 4.7 KB runs code
- scripts/fetch_endpoint.py 8.4 KB runs code
- scripts/fetch_estimates.py 7.1 KB runs code
- scripts/fetch_financials.py 9.5 KB runs code
- scripts/fetch_kline.py 5.6 KB runs code
- scripts/fetch_pe_history.py 4.7 KB runs code
- scripts/fetch_profile.py 7.4 KB runs code
- scripts/fetch_quote.py 8.0 KB runs code
- scripts/fetch_trade_calendar.py 5.0 KB runs code
- scripts/requirements.lock.txt 504 B
- scripts/requirements.txt 185 B
- scripts/sources/__init__.py 380 B runs code
- scripts/sources/_http.py 13 KB runs code
- scripts/sources/baidu.py 1.7 KB runs code
- scripts/sources/baostock_src.py 7.2 KB runs code
- scripts/sources/cboe.py 5.3 KB runs code
- scripts/sources/cls.py 1.3 KB runs code
- scripts/sources/cninfo.py 3.6 KB runs code
- scripts/sources/commodity.py 12 KB runs code
- scripts/sources/datacal.py 7.2 KB runs code
- scripts/sources/eastmoney.py 40 KB runs code
- scripts/sources/exa.py 7.9 KB runs code
- scripts/sources/exchange.py 4.6 KB runs code
- scripts/sources/finra.py 3.0 KB runs code
- scripts/sources/headlines.py 15 KB runs code
- scripts/sources/hiring.py 11 KB runs code
- scripts/sources/indicators.py 6.1 KB runs code
- scripts/sources/industry.py 19 KB runs code
- scripts/sources/iwencai.py 2.2 KB runs code
- scripts/sources/macro_us.py 2.1 KB runs code
- scripts/sources/macro.py 6.4 KB runs code
- scripts/sources/mappers_cn.py 31 KB runs code
- scripts/sources/mappers_commodity.py 5.1 KB runs code
- scripts/sources/mappers_datacal.py 3.7 KB runs code
- scripts/sources/mappers_global.py 31 KB runs code
- scripts/sources/mappers_headlines.py 6.7 KB runs code
- scripts/sources/mappers_hiring.py 4.4 KB runs code
- scripts/sources/mappers_industry.py 9.5 KB runs code
- scripts/sources/mappers_policy.py 5.9 KB runs code
- scripts/sources/mappers_probability.py 8.9 KB runs code
- scripts/sources/mappers.py 34 KB runs code
- scripts/sources/mootdx_src.py 6.0 KB runs code
- scripts/sources/policy.py 21 KB runs code
- scripts/sources/probability.py 34 KB runs code
- scripts/sources/rss.py 5.9 KB runs code
- scripts/sources/sec.py 9.0 KB runs code
- scripts/sources/sina.py 12 KB runs code
- scripts/sources/sw.py 4.9 KB runs code
- scripts/sources/tencent.py 5.4 KB runs code
- scripts/sources/textsafe.py 6.6 KB runs code
- scripts/sources/ths.py 3.5 KB runs code
- scripts/sources/yahoo.py 9.5 KB runs code
- scripts/tests/test_commodity.py 11 KB runs code
- scripts/tests/test_core_retrieval.py 38 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 93 lines · 127 tokens per session scan A dc272d9b7f20
data-access is a skill published in the GitHub repository simonlin1212/Vibe-Research (2,364 stars, last pushed today), licensed MIT. It adds 127 tokens to every session and 4,105 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
daily-deep-brief
8 点这个时点：HK 开盘前 90 分钟，US 已收盘 4 小时。盘前是 deep think 最好的窗口 — 有完整夜间消息面，没有盘中执行压力。.
hk-stock-analysis
Workspace-aware Hong Kong stock analysis for kcn. Routes through clawock analyze-hk (Tencent primary + Eastmoney full-batch independent cross-check/fallback → stooq → yfinance) for price/技术指标/news, layered with HK-specific concepts — 南向资金, HSTECH 方向, 杠杆 ETF 衰减, 老千股警惕, T+0 无涨跌幅. Use when user asks about a HK ticker…
us-stock-analysis
Workspace-aware US stock analysis for kcn. Routes through clawock analyze-us / clawock us-quotes instead of generic web search, then layers fundamental/technical/news analysis on top. Use when user asks to analyze a US ticker (e.g. "analyze AAPL", "look at RKLB", "compare TSLA vs NVDA"), check earnings, run…
portfolio-swarm-review
Multi-agent swarm review of kcn's current holdings. Inspired by TauricResearch/TradingAgents framework already in workspace — three-tier analysis (analysts → bull/bear debate → risk debate + judge) with confidence scoring. Use for post-close reviews, holiday/next-session planning, pre-add sizing decisions, and any…
investment-decision
Run a clawock investment decision — read the prepared request, research with the host's own tools, write decision.json with evidence and an explicit bull/bear debate, and let Python validate and settle. Use when the user asks for an investment decision or a clawock run request is present.
investment-decision
Read the clawock request file, write decision.json, let clawock validate. Use when a clawock run request is present in .clawock/work/.