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/travisun/opptrix/ht-ffscorenpx skills add Travisun/Opptrix --skill ht-ffscoregit clone --depth 1 https://github.com/Travisun/OpptrixWrote 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/travisun/opptrix/ht-ffscore)<a href="https://agentmods.dev/skills/travisun/opptrix/ht-ffscore"><img src="https://agentmods.dev/badge/skills/travisun/opptrix/ht-ffscore.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 | $0.00089 | $0.00957 |
| Opus 5 | $0.00044 | $0.00478 |
| Sonnet 5 | $0.00018 | $0.00191 |
| Haiku 4.5 | $0.00009 | $0.00096 |
Grade A, and why
ht-ffscore 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 4d 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.
What it actually says
华泰 FFScore 财务打分
方法溯源 华泰价值选股 FFScore / 比乔斯基 F-Score；本技能由 Agent 取财务写入工作区后，用自包含脚本打分排序。脚本不联网取数。
何时使用
用户要对一组 A 股做财务质量/改善多维打分排序（FFScore 5 维或 F-Score 9 维）。默认交付网页。
边界：神奇公式双因子用 @skill:lean-magic-formula；一般条件筛选另议。本技能不承诺与研报回测可复现。
非目标
- 不荐股、不给目标价/仓位
- 不调用聚宽/外部行情 SDK
- 不与
lean-magic-formula/lean-qc500-style-screen合并
取数与运行
- 确认宇宙（成分 / 用户清单 /
ask_user）。 - 用
get_instrument_financials（及可得资产负债字段）整理为panels.financials。 workspace_write输入 JSON →opptrix_run：
python scripts/ht_ffscore.py --input data.json --output result.json
- 解读
signal排名表 →list_web_vendor→create_web。
输入 schema（要点）
{
"meta": { "skill": "ht-ffscore", "asof": "2026-08-16" },
"panels": {
"financials": [
{
"symbol": "600519.SH",
"report_date": "2024-12-31",
"roe": 0.25,
"roe_prev": 0.22,
"roa": 12.0,
"operating_revenue": 1e10,
"total_assets": 2e11,
"net_operate_cash_flow": 5e9
}
]
},
"params": { "mode": "ffscore" }
}
params.mode：ffscore（默认，5 维）| fscore（9 维）| both。
缺字段的分项跳过；整表无财务 → ok:false + 说明。
输出解读
signal：按得分降序的 ranking（含分项 components）metrics：样本数、均分等assumptions：口径与缺项声明
依赖
仅 Python 标准库。
网页报告建议目录
- 宇宙与报告期
- 打分规则与字段映射假设
- 排名表（事实）
- 缺项与完整度
- 事实 / 假设 / 推断
- 免责声明（非荐股）
禁止
- 编造财务数字；无
panels.financials仍输出假排名 - 把高分写成「必买」
- 无交付就结束（默认 web）
What ships with it
2 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.
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.
- 4d ago First seen · 99 lines · 89 tokens per session scan A c4d736b0443b
ht-ffscore is a skill published in the GitHub repository Travisun/Opptrix (229 stars, last pushed today), licensed Apache-2.0. It adds 89 tokens to every session and 957 once invoked, about $0.0004 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
national-team-position
估计中国 A 股"国家队"(中央汇金)的宽基 ETF 持仓变动趋势——追踪上交所沪深300/上证50/中证500/中证1000/中证A500/科创50 的 ETF 份额变化,叠加各自指数走势,生成六合一总图与各指数单图。Estimates China's "national team" (Central Huijin) broad-base ETF positioning by tracking Shanghai Stock Exchange ETF share changes. Use when the user asks about 国家队持仓 / 国家队仓位 / 中央汇金持仓 / 国家队加仓 / 国家队减仓 /…
caijing-ipo-hk
财经·港股打新——完整打新顾问 + 发行体检。回答"这只港股新股怎么打、打多少、破发统计风险多大"。内置：可配置用户档案（券商/资金/风险偏好）、主流券商费率预设、申请档位与历史配售证据、绝对孖展金额、因子体检（保荐人/基石/定价四分位/结构/热度/情绪/18A/18C）、三档条件性方案与执行清单。含招股期与数据时效 gate；默认自用模式可给条件性申购建议，对客输出仍按 compliance-rendering 降级。触发："分析这个新股""XX招股了""怎么打""打多少手""港股打新""孖展""暗盘"；命令 /caijing:ipo-hk。.
caijing-fundamental
财经·基本面——产出一份分析师级、全面、详细、前瞻、论点驱动的深度研究文档。对 A股/港股上市公司：立核心论点与分歧视角（市场信什么、我哪里不同），在 2–3 个决定命运的胜负手上量化钻透（带估算区间与置信度），十四段全面框架每段有真分析与子项，配三情景前瞻盈利、估值矩阵、多投资哲学对撞、Pre-Mortem 反偏见与长期跟踪清单。默认产出研报级 docx；客户合规版按需。触发："全面分析XX""XX基本面怎么样""这公司质地如何"；命令 /caijing:fundamental。.
rodya-caijing-studio
财经内容台——面向个人投资者、研究人员与券商投顾/客户经理的 A股+港股专业研究和内容生产工具包。用于全面分析公司基本面、点评财报、判断估值位置、排查财务风险、梳理产业链、体检 A股/港股新股；所有使用者默认获得同一专业研究内核和研报级 docx，客户合规版按需生成，卡片/文案/PPT 按用户点名触发。稳定命令：/caijing:fundamental、/caijing:earnings、/caijing:valuation、/caijing:risk、/caijing:industry、/caijing:ipo-a、/caijing:ipo-hk。.
caijing-earnings
财经·财报——只回答"这季财报行不行/财报前该看什么"。财报发布后的事件驱动快评：最新期核验 gate、同比环比速览、超预期三层基准判定（一致预期优先于公司指引，公司指引优先于历史季节性）、扣非与现金流含金量验证、前瞻指标（合同负债/订单/资本开支）、对投资逻辑支柱的增强/削弱/中性影响；财报发布前可做 earnings preview（关键指标、三情景、催化清单）。默认产出专业版财报点评 docx；客户版财报速览卡+文案按需。触发："XX出财报了""点评XX年报/中报/季报""这季业绩怎么样""财报前看什么""业绩预览"；命令 /caijing:earnings。.
caijing-valuation
财经·估值——只回答"现在贵不贵"。估值锚按行业适配（周期股低PE陷阱/银行PB-ROE/亏损成长PS）、历史分位 band（纵向）、同业四分位对标（横向）、估值中枢有效性检查（价值陷阱提示）、反向 DCF 隐含预期（不给目标价）、三情景变量敏感性。默认产出专业版估值分析 docx；客户版分位 band 卡+文案按需。触发："XX贵不贵""现在什么位置""估值怎么样""能不能买"（只答贵不贵）；命令 /caijing:valuation。.