据 AlphaMoat 统计,截至 2026-10,Odoo Reporting 累计下载 7,415 次、获 8 个星标,按累计下载量在本次参与排名的 10,914 个 AI 技能中排名第 3,783 位,在「数据分析」分类(915 个)中排名第 350 位。
列表默认按累计下载量排序,也可按星标排序;详情页排名按累计下载量计算。
| 平台 | 累计下载 | 星标 | 评论 |
|---|---|---|---|
| ClawHub ↗ | 4,688 | 7 | - |
| SkillHub ↗ | 7,415 | 8 | — |
各平台的计数口径不同,统计时间不一致。
可查销售人员业绩、客户分析、订单、发票、CRM、会计、增值税、库存与应收应付,并生成 WhatsApp 卡片、PDF 和 Excel。写操作在客户端一律被拦,报表只落在本地目录,网络也只连 .env 里配置的 Odoo 地址。
资料发布于 2026-10-27
以下是根据技能文档生成的示例,不代表实际运行结果。
2026-09 起统计方法调整:同名技能按作者分别统计,并覆盖全部技能。图中虚线为调整前的月份,与之后的数据不直接比较,也不计算跨越调整的增长。
| 月份 | 累计下载 | 环比 | 星标 |
|---|---|---|---|
| 2026-10 | 7,415 | — | 8 |
| 2026-09 | 5,214 | — | 8 |
| 2026-08 | 4,657 | +2.8% | 8 |
| 2026-07 | 4,532 | +9.5% | 8 |
| 2026-06 | 4,138 | -3.4% | 7 |
| 2026-05 | 4,283 | — | 7 |
| 2026-03 | 3,955 | — | 7 |
以下是作者发布的原始文件,按纯文本显示,本站未做修改或审核;其中的链接不可点击,命令请核实后再运行。抓取于 2026-10-23。
--- name: odoo description: "Query Odoo data including salesperson performance, customer analytics, orders, invoices, CRM, accounting, VAT, inventory, and AR/AP. Generates WhatsApp cards, PDFs, Excel. Use when user explicitly mentions Odoo or asks for Odoo data." --- # Odoo Financial Intelligence **Read-only, Evidence-First, Ledger-Based Reports** ## Quick Reference: Common Odoo Models
| Model | What It Contains | Use For |
|-------|------------------|---------|
| `res.users` | Users/Salespeople | Find salesperson by name, get user_id |
| `sale.order` | Sales Orders | Revenue by salesperson, order counts, status |
| `account.move` | Invoices/Journal Entries | Invoice tracking, payments, P&L data |
| `res.partner` | Contacts/Customers | Customer info, top customers by revenue |
| `product.product` | Products | Product sales, inventory |
| `account.account` | Chart of Accounts | Financial reporting, balance sheet |
| `account.move.line` | Journal Lines | Detailed ledger entries |
## Security & Credentials
### Required Environment Variables
This skill requires Odoo connection credentials stored in `assets/autonomous-cfo/.env`:
| Variable | Description | Secret |
|----------|-------------|--------|
| `ODOO_URL` | Odoo instance URL (e.g., `https://your-odoo.com`) | No |
| `ODOO_DB` | Odoo database name | No |
| `ODOO_USER` | Odoo username/email | No |
| `ODOO_PASSWORD` | Odoo password or API key | **Yes** |
**Setup:**
```bash
cd skills/odoo/assets/autonomous-cfo
cp .env.example .env
# Edit .env with your actual credentials
nano .env
```
### Model Invocation Policy
**Model invocation is DISABLED** per `skill.json` policy. This skill handles sensitive financial data and external Odoo connections — it must be explicitly invoked by the user.
**Data Handling:** All queries are read-only. No data is modified or exfiltrated.
### Data Handling
- **Read-only:** All mutating methods (`create`, `write`, `unlink`, etc.) are blocked at the client level
- **No exfiltration:** Reports are generated locally in `assets/autonomous-cfo/output/`
- **Network endpoints:** Only connects to the Odoo URL specified in `.env`
- **Output formats:** PDF, Excel, and WhatsApp image cards (local files only)
### Installation
The skill requires a Python virtual environment with specific packages:
```bash
cd skills/odoo/assets/autonomous-cfo
./install.sh
```
Or manually:
```bash
cd skills/odoo/assets/autonomous-cfo
python3 -m venv venv
./venv/bin/pip install -r requirements.txt
```
**Dependencies:** `requests`, `matplotlib`, `pillow`, `fpdf2`, `openpyxl`
## Critical Rules
1. **NEVER assume** - Always ask clarifying questions before generating reports
2. **Multi-company check** - If multiple companies exist, ASK which one to use
3. **Ledger-based** - Use Chart of Accounts and journal entries (account.move.line), not just invoice summaries
4. **Verify periods** - Confirm date ranges with user before running
5. **No silent defaults** - Every assumption must be confirmed
## Before Any Report, Ask:
1. "Which company should I use?" (if multiple exist)
2. "What period? (from/to dates)"
3. "Which accounts or account types to include?"
4. "Any specific breakdown needed?" (by account, by partner, by journal, etc.)
5. "Output format preference?" (PDF, WhatsApp cards, or both)
## Entrypoint
Uses the venv with fpdf2, matplotlib, pillow for proper PDF/chart generation:
```bash
./skills/odoo/assets/autonomous-cfo/venv/bin/python ./skills/odoo/assets/autonomous-cfo/src/tools/cfo_cli.py <command>
```
Or from the skill directory:
```bash
cd skills/odoo/assets/autonomous-cfo && ./venv/bin/python src/tools/cfo_cli.py <command>
```
## Chart of Accounts Based Reporting
Reports should be built from:
- `account.account` - Chart of Accounts structure (code, name, type, internal_group)
- `account.move.line` - Journal entry lines (debit, credit, account_id, date)
- `account.journal` - Source journals (type: sale, purchase, cash, bank, general)
### Account Internal Groups
- **ASSET** - Assets (current, non-current, cash, receivables)
- **LIABILITY** - Liabilities (payables, taxes, accrued)
- **EQUITY** - Owner's equity
- **INCOME** - Revenue accounts
- **EXPENSE** - Cost and expense accounts
- **OFF_BALANCE** - Off-balance sheet accounts
### Common Account Types
- `asset_cash` - Bank and cash accounts
- `asset_receivable` - Accounts receivable
- `asset_current` - Current assets
- `liability_payable` - Accounts payable
- `income` - Revenue
- `expense` - Expenses
### Special Equity Types (Odoo-Specific)
- `equity` - Standard equity accounts (share capital, retained earnings)
- `equity_unaffected` - **Suspense account** for undistributed profits/losses (e.g., 999999)
**CRITICAL for Balance Sheet:**
Odoo's `equity_unaffected` is a SUSPENSE account. Do NOT use its ledger balance directly.
**Correct Equity Calculation:**
1. **Equity Proper** (type: `equity`) - Use ledger balance (credit - debit)
2. **Retained Earnings** (prior years) - Ledger balance from `equity_unaffected`
3. **Current Year Earnings** - Compute real-time: Income - Expenses
```
Total Equity = Equity Proper + Retained Earnings + Current Year Earnings
```
Where Current Year Earnings = Σ(income credit-debit) - Σ(expense debit-credit)
**Why this matters:** Odoo computes Current Year Earnings in real-time on the Balance Sheet. Using only the `equity_unaffected` ledger balance will cause the balance sheet to NOT balance.
## Automatic Reporting Standard Detection
The skill automatically detects the company's accounting standard based on country/jurisdiction and formats reports accordingly.
**Supported Standards:**
| Standard | Jurisdiction | Notes |
|----------|--------------|-------|
| IFRS | International | Default for most countries |
| US GAAP | United States | SEC registrants |
| Ind-AS | India | Indian GAAP converged with IFRS |
| UK GAAP | United Kingdom | FRS 102 |
| SOCPA | Saudi Arabia | IFRS adopted |
| EU IFRS | European Union | IAS Regulation |
| CAS | China | Chinese Accounting Standards |
| JGAAP | Japan | Japanese GAAP |
| ASPE | Canada | Private enterprises |
| AASB | Australia | Australian standards |
**Detection Logic:**
1. Query company's country from `res.company`
2. Map country code to reporting standard
3. Apply standard-specific formatting:
- Number format (1,234.56 vs 1.234,56)
- Negative display ((123) vs -123)
- Date format (DD/MM/YYYY vs MM/DD/YYYY)
- Statement titles (Balance Sheet vs Statement of Financial Position)
- Cash flow method (indirect vs direct)
**Override:**
```python
# Force a specific standard
reporter.generate(..., standard="US_GAAP")
```
## Commands
### Sales & CRM Queries
```bash
# Salesperson performance - use direct RPC for flexibility
./venv/bin/python -c "
from src.visualizers.whatsapp_cards import WhatsAppCardGenerator
# Query sale.order by user_id, aggregate by month/status
# Generate cards with generate_kpi_card() and generate_comparison_card()
"
# Example RPC query for salesperson:
# - sale.order (user_id, amount_total, state, date_order)
# - account.move (invoice_user_id, amount_total, payment_state)
# - res.users (salesperson info)
# - res.partner (customer info)
```
### Pre-built Reports
```bash
# Financial Health - cash flow, liquidity, burn rate, runway
cfo_cli.py health --from YYYY-MM-DD --to YYYY-MM-DD --company-id ID
# Revenue Analytics - MoM trends, top customers
cfo_cli.py revenue --from YYYY-MM-DD --to YYYY-MM-DD --company-id ID
# AR/AP Aging - overdue buckets
cfo_cli.py aging --as-of YYYY-MM-DD --company-id ID
# Expense Breakdown - by vendor/category
cfo_cli.py expenses --from YYYY-MM-DD --to YYYY-MM-DD --company-id ID
# Executive Summary - one-page CFO snapshot
cfo_cli.py executive --from YYYY-MM-DD --to YYYY-MM-DD --company-id ID
```
### Direct RPC Queries (Advanced)
For sales/CRM data not covered by pre-built commands, use direct RPC:
```python
# Query sales orders by salesperson
orders = jsonrpc('sale.order', 'search_read',
[[('user_id', '=', SALESPERSON_ID)]],
{'fields': ['name', 'partner_id', 'amount_total', 'state', 'date_order']})
# Query invoices by salesperson
invoices = jsonrpc('account.move', 'search_read',
[[('invoice_user_id', '=', SALESPERSON_ID), ('move_type', '=', 'out_invoice')]],
{'fields': ['name', 'partner_id', 'amount_total', 'payment_state']})
# Find salesperson by name
users = jsonrpc('res.users', 'search_read',
[[('name', 'ilike', 'name_here')]],
{'fields': ['id', 'name', 'login']})
```
### Ad-hoc Reports
```bash
# Custom comparison
cfo_cli.py adhoc --from YYYY-MM-DD --to YYYY-MM-DD --metric-a "revenue" --metric-b "expenses"
# Examples:
cfo_cli.py adhoc --metric-a "cash in" --metric-b "cash out"
cfo_cli.py adhoc --metric-a "direct expenses" --metric-b "indirect expenses"
```
### Output Formats
```bash
--output whatsapp # Dark theme 1080x1080 PNG cards
--output pdf # Light theme A4 PDF
--output excel # Excel workbook (.xlsx)
--output both # PDF + WhatsApp cards
--output all # PDF + Excel + WhatsApp cards
```
## Automatic Visualizations
**Reports always include appropriate visualizations by default:**
| Report | Auto-Included Charts |
|--------|---------------------|
| Financial Health | Cash position, burn rate trend, runway |
| Revenue | MoM trend, top customers, growth KPI |
| AR/AP Aging | Aging buckets pie, overdue highlights |
| Expenses | Category breakdown, trend, top vendors |
| Executive | All KPI cards, summary charts |
| Balance Sheet | Asset/liability composition |
| P&L | Revenue vs expense, margin trend |
| Cash Flow | Operating breakdown, cash trend |
**Rule:** If visualization makes the report clearer, include it automatically. Never ask "do you want charts?" — just add them.
## Interactive Param Collection
If required params are missing, the skill will ask:
1. **Company:** "Which company?" (list available options)
2. **Period:** "What period? (e.g., 'last month', 'Q4 2024', custom dates)"
3. **Accounts:** "Which accounts or groups?" (e.g., 'all income', 'bank accounts only')
4. **Breakdown:** "Group by? (Month, Customer, Category, Account)"
5. **Output:** "Output format? (WhatsApp cards, PDF, Both)"
## How to Use in Chat
Just ask naturally:
**Sales & CRM:**
- "How is [name] salesperson performance?"
- "Show me top customers for [salesperson]"
- "Compare sales team performance"
- "Which salesperson has the most orders?"
**Financial Reports:**
- "Give me a financial health report for last quarter"
- "Show revenue vs expenses for the past 6 months"
- "What's my AR aging?"
- "Generate an executive summary for this month"
- "Show me profit & loss statement based on chart of accounts"
**General Queries:**
- "How many orders did we get this month?"
- "Who are the top 10 customers?"
- "Show invoice status for [customer name]"
The skill will:
1. Check for multiple companies and ask which one
2. Parse your request
3. Ask for any missing info
4. Fetch data from Odoo using ledger entries or direct RPC
5. Generate charts + WhatsApp cards
6. Deliver via WhatsApp cards and/or PDF
## Hard Rules
1. Odoo RPC output is source of truth
2. Strict read-only (no create/write/unlink)
3. No proactive actions unless requested
4. Every number includes methodology note
5. Always verify with user before assuming
6. **Always include visualizations** - If a report benefits from charts/graphs, include them automatically without asking. Reports should be visually complete.
## Diagnostics
```bash
python3 ./skills/odoo/assets/autonomous-cfo/src/tools/cfo_cli.py doctor
```
## Report Themes
- **WhatsApp Cards:** "Midnight Ledger" — Navy-black (#0a0e1a), copper glow (#cd7f32)
- **PDF Reports:** Clean white, copper accents, professional layout
中国金融机构数据爬取与分析综合 Skill v12.0.0。2334家/43大类机构名单、60+财经媒体源舆情、 A股报告/公告/研报、60+种查询多源金融数据(自动降级+缓存)、爬取调度与网页操作增强、 实时数据中枢、24 技术指标引擎、多源证据融合、110 MCP 工具 + 10 Resource + 14 Prompt。 🆕 v12.0.0 四条线(信息获取 / 打包下载 / 数据分析 / 登录): 编排:gather_info() 一次跑完"路由→抓取→去占位→质量分→能力边界→回执→打包", 8 类意图各带问法、输入输出、典型耗时与 not_for(说清哪件事做不了)。 诚实:反爬占位条目一律剔除并记账,空结果报 E_SOURCE_EMPTY,不把假数据判成 B 级成功。 交付:包内自带 manifest.json(逐文件 sha256),verify 查得出改动/缺失/多塞, 批量下载单条失败不中断整批,失败清单可直接只重下那几条。 登录:Cookie 由用户导出或扫码接管,档案化 + 健康度四档,对外输出全脱敏(值永不回显)。 何时触发:爬取机构、舆情爬取、东方财富、公告下载、实时行情、技术指标、统一查询、 交付包、批量重下、我已经登录了、登录态还有效吗(完整触发词见下方 auto_trigger)。
大宗商品价格数据查询技能。用于查询大宗商品期现货价格、宏观经济指标、产业链供需数据、进出口库存数据、金融市场数据等。当用户询问:(1) 期货/现货价格查询,如"螺纹钢价格"、"铜期货收盘价"; (2) 历史行情或趋势分析;(3) 区域或企业价格对比;(4) 宏观经济指标,如GDP、CPI;(5) 产业链上下游数据;(6) 多市场金融数据(期权、股票、债券、外汇);(7) 结算确认单生成时触发此技能。
32种图表,3 个主题,2步生成: 1.上传数据 — 将 CSV / Excel / JSON 等格式的数据文件拖入对话框; 2.查看结果 — 交互式图表(HTML),可保存图表为图片.
AdMapix 的取数层:把广告与应用相关的查询转成接口调用,只回原始 JSON。
论文配图一键生成:一条命令从 JSON/CSV/Excel 生成顶刊级学术图表,科研绘图 与期刊配图全场景覆盖。40 种图表(森林图/KM 生存曲线(自动风险表)/竞争风险累计发生率/ROC/校准 曲线/决策曲线 DCA/列线图/RCS 非线性剂量反应/亚组分析森林图/RMST 限制平均生存时间/Galbraith 径向图/火山图/雨云图/山脊图/哑铃图/游泳图(逐例时长+里程碑)/ 肿瘤缓解瀑布图/UpSet 交集/PRISMA 流程图/漏斗/BA 一致性/PCA/热图/箱线/小提琴/ 组合图/Cox 多因素回归森林图等),9 套配色(含色盲安全 Okabe-Ito 与期刊色板),9 种期刊投稿预设, YAML 流水线一条命令出全套论文图,一次渲染多格式导出(TIFF/PNG/PDF), PDF 重叠门禁,600dpi。纯本地运行,数据不出机,中文零配置。适用:写论文/ 投稿要配图、把数据变成可投稿图表、批量重出全套图。不适用:文本润色/翻译/ 改写(请用写作类工具);商业报表/仪表盘/海报装饰;无数据支撑的纯示意图。
Intelligent chart generation and data analysis skill. Reads user-supplied data files (CSV/Excel/JSON), analyzes data characteristics with LLM assistance, auto-recommends and generates interactive ECharts visualizations. Use when the user asks to analyze data, generate charts, create visualizations, or work with tabular data files.
全球智能投研工具(stock-researcher)v13.0:多源数据 + 概率预测 + 可导出报告。 核心能力: - 多源数据路由:按市场/类别选源,逐级降级;每个结果标注来源与时效等级;源不可用时明确标 unavailable - 全资产量化:股票/ETF/场外基金/指数/期货(期限结构、基差、持仓、折溢价)多因子分析 - 概率预测:1d/5d/1M/3M/6M 经验分布 + 分位数 + 方向概率 + 区间,kNN 邻域收缩融合全局分布; 样本不足自动弃权,可按历史命中样本(同桶)做概率校准 - 证据链简报:每条结论绑定来源、时间与数据质量,冲突信号保留而不抹平 - 报告导出:Word/PPT/PDF/Markdown/JSON/纯文本,纯标准库渲染 + 可选第三方引擎 - MCP Server 24 工具,Claude Code 可直接调用;知识库 12 适配器双向同步 三条铁律: - 数据不装:拿不到就报不可用,不拿 0 价、不拿空表冒充 - 预测不装:概率来自历史相似条件的经验分布,样本不足就说不足,不硬给方向 - 评分≠盈利保证:任何结论都带免责声明
走 Tavily 检索公开网页,返回带链接与摘要的结构化结果。
数据月份:2026-10 · 下载/星标数聚合自公开技能注册库(ClawHub、SkillHub),按月更新。口径说明见数据方法。