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/rollysys/use_cninfo/skillnpx skills add rollysys/use_cninfo --skill skillgit clone --depth 1 https://github.com/rollysys/use_cninfoWrote 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/rollysys/use_cninfo/skill)<a href="https://agentmods.dev/skills/rollysys/use_cninfo/skill"><img src="https://agentmods.dev/badge/skills/rollysys/use_cninfo/skill.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 | $0.00139 | $0.01763 |
| Opus 5 | $0.00069 | $0.00881 |
| Sonnet 5 | $0.00028 | $0.00353 |
| Haiku 4.5 | $0.00014 | $0.00176 |
Grade A, and why
cninfo 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
cninfo skill — A 股公告按需抓取
本 skill 是对 cninfo-cli 命令的入口,把巨潮资讯网的所有公开公告(法定披露)按需变成本地 PDF + 全文 md。纯按需,不做后台调度。
触发场景
- "拿 600519 的 2024 年报"
- "茅台 2024 年第一季度报告"
- "贵州茅台最近 30 天的公告"
- "4/29 全市场所有年报"
- "搜含'减持'的公告 / 找重大资产重组类公告"
前置依赖(用户机器上必须先装好)
# 1. 本工具
git clone https://github.com/rollysys/use_cninfo.git
cd use_cninfo && uv tool install -e . # 装上 `cninfo` 命令
# 2. (可选)公告分类器 — 提供 search 的 --type/--sub-type 标签精准过滤
git clone https://github.com/rollysys/announcement_filter.git
cd announcement_filter && pip install -e .
不装 announcement_filter 时:cninfo search --type ... --sub-type ... 会报错并提示安装命令;其它 5 个命令不受影响。
缓存
默认 ~/.cache/cninfo/(可被 $CNINFO_CACHE_DIR 覆盖):
<root>/
├── orgid_map.json # secCode → orgId
├── pdf/<ts_code>/<ann_date>__<ann_id>.pdf
├── md/<ts_code>/<ann_date>__<ann_id>.md # 含 YAML frontmatter
└── meta/<ann_id>.json
命中即秒级返回,不发网络。强制重下用 --force。
6 个命令(给 agent 自己选)
1. 单股指定定期报告(最常用)
cninfo fetch-report 600519 --year 2024 --kind annual
# kind: annual / q1 / h1 / q3
# 输出 JSON: {ann_id, ann_date, title, pdf_path, md_path, total_pages, extracted_pages, text_chars, cache_hit}
板块(sz/sh/bj)默认按代码推测;把握不准时显式 --plate sh。
2. 单股时间窗全公告
cninfo fetch-stock 600519 --since 2024-04-01 --until 2024-04-30 # 仅列表
cninfo fetch-stock 600519 --since 2024-04-01 --until 2024-04-30 --download # 同时落 PDF+md
cninfo fetch-stock 600519 --since 2024-04-01 --until 2024-04-30 --json # JSON 输出
3. 全市场切片
# 4/29 当天 sz plate 全部年报
cninfo list --plate sz --category annual --date 2024-04-29
# 时间窗 + 单只关键词
cninfo list --plate sh --since 2024-04-01 --until 2024-04-30 --keyword "重大资产重组"
# 加 --download 才下载;--limit 50 限制条数
4. 标签搜索(走 announcement_filter)
# 单股 + 时间窗 + 标签
cninfo search --stock 600519 --since 2024-01-01 --until 2024-12-31 \
--type shareholder --sub-type reduce_plan
# 全市场 + 单日 + 标签
cninfo search --plate sz --date 2024-04-29 --type shareholder --sub-type reduce_plan
# 不带标签时退化为 cninfo searchkey 模糊匹配 title
cninfo search --plate sh --date 2024-04-29 --keyword "重组"
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 · 130 lines · 139 tokens per session scan A be9d54626b9a
cninfo is a skill published in the GitHub repository rollysys/use_cninfo (22 stars, last pushed 4mo ago), licensed MIT. It adds 139 tokens to every session and 1,763 once invoked, about $0.0007 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
sector-rotation
行业轮动分析——申万行业景气度评分、行业动量排名、产业链传导、估值/盈利/资金流多维比较框架.
twitter-reader
Read Twitter/X for financial research using opencli (read-only). Use this skill whenever the user wants to read their Twitter feed, search for financial tweets, view bookmarks, look up user profiles, or gather market sentiment from Twitter/X. Triggers include: "check my feed", "search Twitter for", "show my…
strategy-pivot-designer
Detect backtest iteration stagnation and generate structurally different strategy pivot proposals when parameter tuning reaches a local optimum.
chenhao-limit-up
Use when evaluating A-share limit-up (涨停板) setups through Chen Hao's sentiment and momentum lens: market emotion cycles, board strength, follow-through, and short-term aggressive momentum trading.
trading-risk-gate
Unified pre-trade safety gate: Ruin check (Law #1), ergodicity audit, and win-rate dominance validation. Absorbs: ergodicity-check, law-of-ruin, win-rate-dominance.
vectorbt
High-performance vectorized backtesting with parameter optimization, portfolio simulation, and rich performance metrics.