AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Cue Financial Deep Read

skill-sensedeal-cue-skills-cue-financial-deep-read · by sensedeal

>

No reviews yet
0 installs
33 views
0.0% view→install

Install

$ agentstack add skill-sensedeal-cue-skills-cue-financial-deep-read

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-sensedeal-cue-skills-cue-financial-deep-read)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Cue Financial Deep Read? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Cue「财报深读」研究 skill

加载本 skill 后,你可以用 Cue 跑这个场景的深度研究(多源公开数据交叉、结论带来源链接)。

何时用

财报深读:财务与业绩、信披与监管。

当前可用搭子(仅供理解;运行时以 live 为准)

  • 上市公司财报分析:围绕指定上市公司最新一期财报,从核心数据变动、业务驱动因子、利润含金量、产业链话语权与典型财务信号产出一份带可回查出处的
  • 财报极速排雷:用财务诊断框架快速扫描企业利润含金量、产业链话语权、典型财务雷区,叠加司法监管与同业对照,产出可回查的风险信号清单与证据
  • 财务质量与供应链核查:穿透三张报表、经营指标、客户供应商披露与应收款账龄,区分报表事实与机构解读,产出财务质量与供应链风险证据框架。
  • 财报前瞻指引兑现核查:管理层的业绩指引,到底能不能信?把公司的业绩预告与发布时的市场一致预期、以及后续实际兑现三方对账,看它是连续兑现还是惯于
  • 财报附注雷达:给定一个风险信号或主题,跨全 A 股年报附注做语义检索,定位讨论它的公司清单、章节摘要与页码溯源,并归纳共性诱因——一句
  • 可转债与审计意见核查:交叉核验可转债基本信息、正股、审计意见与全市场清单,快速确认转债主体与财务审计质量,产出可复核的核查底稿。

准备 Cue runner(首次用时,幂等)

本 skill 不自带脚本,靠 Cue 开源 runner 跑研究。先确认 runner 是否就绪:

  • 若你已安装 cue-skills(或本 skill 来自整包发布)→ 直接用其中的 cue-research/scripts/research_run.py跳过本节
  • 否则克隆开源仓(含 cue-research + cue-buddy 全套依赖),有则更新、无则克隆(GitHub 不通走镜像):

``bash if [ -d ~/.cue/cue-skills/.git ]; then git -C ~/.cue/cue-skills pull --ff-only else git clone https://github.com/sensedeal/cue-skills ~/.cue/cue-skills \ || git clone https://gitee.com/sensedeal/cue-skills ~/.cue/cue-skills fi ` 之后 runner = ~/.cue/cue-skills/cue-research/scripts/research_run.py。需 git + python3`(runner 仅用标准库)。

怎么跑(搭子是动态的,运行时查 live)

  1. 拉本场景当前搭子GET https://cuecue.cn/api/playbook,找 secondary_category == "财报深读" 的 scene,读 buddies[](每个有 template_id/title/goal)。若该场景当前不在返回里(临时未达展示门槛)→ 告知用户暂不可用。
  2. 选一个搭子委托 cue-research 的匹配逻辑(其 +match/Stage-2:对 goal 做语义匹配、把用户的具体主体从匹配中剥离、弱命中先列 ≤2 候选确认)——不要只按字面 title 关键词裸选。取选中搭子的 template_id
  3. 确认 credits(强制):跑深度研究消耗 credits。运行前显式问用户「将用搭子 X 跑【主体】,耗 credits,是否继续?」并等确认。
  4. python3 ~/.cue/cue-skills/cue-research/scripts/research_run.py --query "" --template-id (用上一节就绪的 runner 路径;已装 cue-skills 则用你本地的 cue-research/scripts/research_run.py)。深度研究 3–15 分钟;长跑 live 流常不带报告段,用 replay 取最终报告。 读 runner 末行 RESULT ok|emptyempty → 告知用户本次未取到内容、可换主体/搭子重试,不要编造
  5. 回报:把带来源链接的报告交给用户,不去掉来源、不杜撰。

前置

  • Cue 账号 API key(cue CLI 登录后在 ~/.cue/config.json,runner 自动读);新账号送免费积分(注册 50 + 每天 10),可先免费试。
  • git + python3(自举 runner 用;runner 仅标准库)。
  • 跑深度研究消耗 credits;只覆盖公开数据,不替代尽调/法律/核保。

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.