Agent skill

Financial Expense Automation

by DjangoPeng in DjangoPeng/agentic-ai

当用户上传 PDF、JPG、JPEG、PNG 附件,或消息中包含"报销""发票""票据""录入""火车票""机票"等关键词时触发。对附件内容进行识别,若确认为报销票据则提取结构化字段并写入飞书多维表格;若识别后内容不是报销票据,告知用户并终止流程。

MITAuto-check passedDocuments & Office

Install Financial Expense Automation

skills CLI
$ npx skills add DjangoPeng/agentic-ai --skill financial-expense-automation -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install DjangoPeng/agentic-ai financial-expense-automation --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/DjangoPeng/agentic-ai.git skills-src && mkdir -p .claude/skills && cp -r skills-src/financial-automation/skills/financial-expense-automation .claude/skills/financial-expense-automation && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
financial-expense-automation
GitHub stars
152
Token cost
~816 tokens
SKILL.md length
257 words
Files
2
Skills in repo
5
Repo updated
First seen
Licence
MIT

At a glance

当用户上传 PDF、JPG、JPEG、PNG 附件,或消息中包含"报销""发票""票据""录入""火车票""机票"等关键词时触发。对附件内容进行识别,若确认为报销票据则提取结构化字段并写入飞书多维表格;若识别后内容不是报销票据,告知用户并终止流程。

  • Works in 3 steps: 环境变量 FINANCIAL_AUTOMATION_ROOT → ~/projects/agentic-ai/financial-automation → ~/.openclaw/workspace/financial-automation
  • Tasks that involve PDF
  • SKILL.md covers 目标, 项目依赖, 支持输入 and 非报销内容处理, plus 11 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Financial Expense Automation is an agent skill from DjangoPeng/agentic-ai. 当用户上传 PDF、JPG、JPEG、PNG 附件,或消息中包含"报销""发票""票据""录入""火车票""机票"等关键词时触发。对附件内容进行识别,若确认为报销票据则提取结构化字段并写入飞书多维表格;若识别后内容不是报销票据,告知用户并终止流程。

Its SKILL.md is about 820 tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).

It sits in Documents & Office, covering PDF. It works with Feishu (Lark). The repository describes itself as: 一套经过生产验证的 OpenClaw + Claude Code 实战知识库:涵盖生产部署、IM 接入、模型配置、安全加固,以及一系列可落地的 AI Agent 业务流项目(小红书发布 / 财务票据 / 智能早报 / CRM / 量化投研 / 密钥巡检自愈)。 The licence is MIT.

When your agent uses it

  • Tasks that involve PDF

Example prompts

  • “/financial-expense-automation”

Requirements

  • Python 3

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. 环境变量 FINANCIAL_AUTOMATION_ROOT
  2. ~/projects/agentic-ai/financial-automation
  3. ~/.openclaw/workspace/financial-automation

What it can do on your machine

Read from SKILL.md and the folder at commit c8bc8f1. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are python).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Financial Expense Automation loads about 816 tokens when it runs. Until then it costs about 38 tokens; SKILL.md has 257 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~38
When it runs · the whole SKILL.md, loaded when a task matches
~816

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from DjangoPeng/agentic-ai at commit c8bc8f1, republished under its MIT licence (© DjangoPeng). 257 words, ~816 tokens.

Download SKILL.mdSave it as .claude/skills/financial-expense-automation/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
financial-expense-automation
description
当用户上传 PDF、JPG、JPEG、PNG 附件,或消息中包含"报销""发票""票据""录入""火车票""机票"等关键词时触发。对附件内容进行识别,若确认为报销票据则提取结构化字段并写入飞书多维表格;若识别后内容不是报销票据,告知用户并终止流程。

Financial Expense Automation

目标

运行本地 Financial Automation 流水线,对用户上传的报销票据进行:

  • 识别
  • 结构化提取
  • 校验
  • 真实写入飞书多维表格

请始终记住:

  • bitable_write_plan 只是中间产物
  • 识别成功不等于任务完成
  • 只有真实调用 Feishu Bitable create/update 成功,并完成回读确认,才算完成

项目依赖

这个 skill 依赖完整项目仓库,不能只拷贝 SKILL.md 单独使用。

按以下顺序定位项目根目录:

  1. 环境变量 FINANCIAL_AUTOMATION_ROOT
  2. ~/projects/agentic-ai/financial-automation
  3. ~/.openclaw/workspace/financial-automation

若这些路径都不存在,应明确告诉用户:当前环境尚未部署完整项目仓库,请先完成部署。

主要入口与配置:

  • <repo_root>/src/skill_entry.py
  • <repo_root>/config/app_config.yaml

支持输入

附件输入格式:

python
[
    {"file_name": "hotel_invoice.pdf", "content_bytes": b"..."},
    {"file_name": "ticket.jpg", "source_path": "/path/to/ticket.jpg"},
]

支持文件类型:

  • .pdf
  • .jpg
  • .jpeg
  • .png

若过滤后没有可处理附件,应直接告知用户:没有收到可处理的报销附件。

非报销内容处理

识别完成后,若内容不是报销票据(如普通图片、截图、合同等),应:

  1. 告知用户:该附件不是可识别的报销票据
  2. 简述识别到的内容类型
  3. 终止流程,不继续写表

唯一入口

必须通过:

python
from src.skill_entry import run_skill_job
result = run_skill_job(attachments)

如有需要可显式传配置:

python
result = run_skill_job(
    attachments,
    config_path=f"{repo_root}/config/app_config.yaml",
)

不要手工拼接 ingest / OCR / validate / formatter 流程。

正式执行流程

  1. 定位 repo root
  2. 将用户上传文件整理成 run_skill_job(...) 所需的附件 payload
  3. 调用 run_skill_job(...)
  4. 使用返回的 skill_result 作为识别结果主对象
  5. 生成真实写表输入
  6. 若当前会话具备 Feishu Bitable 工具能力,继续执行真实写表
  7. 写入后回读确认,再向用户回复结果

写表强制规则

  1. bitable_write_plan 只是中间产物,不是最终结果
  2. 只要当前会话可用飞书多维表格工具,就必须继续真实写表
  3. 禁止停留在“建议写入 / 准备写入 / 可写入”状态
  4. 只有真正调用 create/update 成功,才算完成
  5. 如果没有真实写入成功,必须明确说明失败点
  6. 禁止把“已识别 / 已生成 plan / 已生成 handoff”描述成已经完成落表

目标表路由规则

  • transportation_fee → 交通报销表
  • 其他费用类票据 → 费用报销表

写入策略

默认采用:

  • update_first_blank_row_then_create

具体规则:

  1. 先查询目标表
  2. 若存在可复用空白行(优先判断 doc_id 为空),优先 update
  3. 若不存在可复用空白行,再 create
  4. 不要盲目追加新记录

附件写入规则

附件字段必须遵守以下规则:

  1. 禁止直接使用通用 Drive upload token 作为 Bitable 附件
  2. 必须先上传到当前 bitable attachment context
  3. 再将返回的合法 file_token 写入附件字段
  4. 图片走 bitable_image
  5. PDF/其他文件走 bitable_file
降级规则

如果当前环境附件链路不可用:

  • 不要伪造 file_token
  • 不要把通用 Drive token 冒充 bitable 附件
  • 可以先真实写入非附件字段
  • 并明确告诉用户:附件尚未成功挂载

注意:

  • 如果附件失败但非附件字段已落表,必须把两件事分开说清楚
  • 如果连真实 create/update 都没发起,也必须明确说明尚未完成真实写表

PDF / 图片识别规则

  • 图片走 OCR
  • PDF 优先走原生文本抽取
  • 如果 PDF 原生抽取结果不足或不可用,应继续走 OCR fallback
  • 不能因为某个提取分支失败就直接把整单描述成“不可识别”

返回对象重点字段

最重要字段:

  • user_summary
  • summary
  • highlights
  • documents
  • review_queue
  • job
  • bitable_write_plan

说明:

  • documents 是标准化结构化结果
  • review_queue 表示需要人工复核的项目
  • bitable_write_plan 是真实写表的中间输入,不是完成态

完成标准

以下情况才算完成:

  • 已成功识别票据
  • 已成功判断目标表
  • 已真实调用 Feishu Bitable create/update
  • 已明确回读或确认写入结果

以下情况都不算完成:

  • 只输出结构化字段
  • 只输出 bitable_write_plan
  • 只说“待写入 / 准备写入 / 建议写入”
  • 只生成 handoff/prompt 但没有真实 create/update

回复用户规则

回复用户时:

  1. 先用 user_summary.headline 概括结果
  2. 若有 review_queue,明确列出复核项与原因
  3. 发票场景总结购方、销方、金额、项目名称等关键字段
  4. 火车票场景总结购买方、路线、日期、乘客、席位等关键字段
  5. 不要只汇报识别结果,必须汇报真实写表结果,或明确失败点

当前业务范围

当前优先支持:

  • 普通电子发票
  • 铁路电子客票

重点提取字段包括:

  • 发票号码
  • 开票日期
  • 金额
  • 币种
  • 购方与销方信息
  • 行项目 / 税率 / 税额
  • 铁路票路线 / 车次 / 乘客 / 出行日期 / 席位
  • 校验结论与复核原因

© DjangoPeng, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in financial-automation/skills/financial-expense-automation of DjangoPeng/agentic-ai.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit c8bc8f1

Compare with similar skills

Financial Expense Automation next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Financial Expense Automation compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Financial Expense Automation this skillDjangoPeng/agentic-ai152—~816Automated safety check: PassMIT
Readninehills/skills280—~2.1kAutomated safety check: WarnNone
Read URLs and PDFstw93/Waza7.2k—~1.8kAutomated safety check: PassMIT
MarkitdownImCa0/just-laws78114 repos~3.2kAutomated safety check: NotesMIT
Gzh Designisjiamu/gzh-design-skill4k—~2.2kAutomated safety check: PassAGPL-3.0
GenOffice Document CLIgenspark-ai/genoffice9.2k—~19kAutomated safety check: PassApache-2.0

Similar skills

  • Read

    ninehills/skills

    Reads URLs and PDFs by fetching source content, defaulting to concise summaries for plain read requests and clean Markdown when asked to convert, save, quote, cite, or feed downstream work.

    280 GitHub stars~2.1k tokensUpdated 3 mo ago
    Documents & OfficeAuto-check: warnings
  • Fetches web pages and PDFs and returns a source-grounded summary, clean Markdown, quotes or citations, routing each kind of link to a suitable fetch method.

    7.2k GitHub stars~1.8k tokensUpdated today
    Documents & OfficeAuto-check passed
  • Markitdown

    ImCa0/just-laws

    Convert files and office documents to Markdown. An agent skill from ImCa0/just-laws.

    781 GitHub starsUsed in 14 repos~3.2k tokens
    Documents & OfficeAuto-check: notes
  • Gzh Design

    isjiamu/gzh-design-skill

    微信公众号文章排版引擎,将 Markdown 转换为可直接粘贴到公众号编辑器的 HTML。主题风格从 references/theme-index.md 注册的自定义主题库中选取,自动章节编号、关键词下划线标记、引言卡片、目录导航、代码块、图片/GIF、作者签名。支持 Markdown / Word(.docx) / PDF / 纯文本输入(非 Markdown…

    4k GitHub stars~2.2k tokensUpdated 2 days ago
    Documents & OfficeAuto-check passed
  • GenOffice Document CLI

    genspark-ai/genoffice

    Creates, converts, reads and edits real pptx, xlsx, docx and PDF files locally through the genoffice command line.

    9.2k GitHub stars~19k tokensUpdated yesterday
    Documents & OfficeAuto-check passed
  • Harness Book Best Practice

    wquguru/harness-books

    Best practices for working on the Harness books repo. An agent skill from wquguru/harness-books.

    3.2k GitHub stars~4.1k tokensUpdated 5 mo ago
    Documents & OfficeAuto-check passed

More from DjangoPeng/agentic-ai

  • 当用户希望生成 AI 早报、运行晨报链路、查看今日资讯摘要,或消息中包含”早报””晨报””今日资讯””新闻摘要”等关键词时触发。

    152 GitHub stars~974 tokensUpdated 3 mo ago
    Auto-check: notes
  • Xhs Auto Publisher

    DjangoPeng/agentic-ai

    面向云服务器的小红书图文自动发布 Skill。适用于需要在 Linux 云服务器上,通过 Playwright/CDP、二维码人工接管、登录缓存、龙虾代发飞书群图片消息、截图留痕与审计日志来完成小红书草稿或发布流程的场景。

    152 GitHub stars~570 tokensUpdated 3 mo ago
    Auto-check passed
  • CRM Assistant

    DjangoPeng/agentic-ai

    将销售会议 transcript、飞书会议原始 JSON、飞书云文档正文或 Word/DOCX 会议纪要,转换成 CRM 结构化结果,并按客户信息表 + 商机推进快照表两表模型生成/同步飞书多维表格记录时使用。适用于多轮客户推进、客户画像增量更新、商机阶段判断、Lead Score 计算,以及“弱值不覆盖旧值、沟通风格/风险顾虑合并”的客户字段更新规则。

    152 GitHub stars~681 tokensUpdated 3 mo ago
    Auto-check passed
  • GitHub Secret Auditor

    DjangoPeng/agentic-ai

    当用户希望让 OpenClaw 通过 ACP 调度 Claude Code,对 GitHub 仓库进行 API Key、Token、密码、私钥、Webhook URL 等敏感信息泄露巡检、自动修复、验收、推送修复 commit,并通过飞书发送巡检报告时使用。

    152 GitHub stars~2.8k tokensUpdated 3 mo ago
    Auto-check: notes

Works with

Questions about Financial Expense Automation

What does Financial Expense Automation do?

当用户上传 PDF、JPG、JPEG、PNG 附件,或消息中包含"报销""发票""票据""录入""火车票""机票"等关键词时触发。对附件内容进行识别,若确认为报销票据则提取结构化字段并写入飞书多维表格;若识别后内容不是报销票据,告知用户并终止流程。. Financial Expense Automation is an agent skill from DjangoPeng/agentic-ai.

When should I use Financial Expense Automation?

Financial Expense Automation fits situations like: tasks that involve PDF.

How do I install Financial Expense Automation in Claude Code?

Run `npx skills add DjangoPeng/agentic-ai --skill financial-expense-automation -a claude-code`. Or copy the skill folder (financial-automation/skills/financial-expense-automation in DjangoPeng/agentic-ai) into .claude/skills/financial-expense-automation in your project. Claude Code loads it when a task matches its description.

How do I install Financial Expense Automation in Codex?

Run `npx skills add DjangoPeng/agentic-ai --skill financial-expense-automation -a codex`. Or copy the skill folder (financial-automation/skills/financial-expense-automation in DjangoPeng/agentic-ai) into .agents/skills/financial-expense-automation in your project. Codex loads it when a task matches its description.

Can I use Financial Expense Automation in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add DjangoPeng/agentic-ai --skill financial-expense-automation -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/financial-expense-automation, .gemini/skills/financial-expense-automation, .github/skills/financial-expense-automation and .opencode/skills/financial-expense-automation in your project.

What does Financial Expense Automation need to run?

SKILL.md names no scripts, command-line tools or credentials: Financial Expense Automation is instructions for the agent only. Our summary lists: Python 3.

Does Financial Expense Automation access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Financial Expense Automation safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Financial Expense Automation use?

Financial Expense Automation is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Financial Expense Automation use?

About 816 tokens (SKILL.md is roughly 3.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Financial Expense Automation?

Skills that share tags, products or a category with Financial Expense Automation: Read (ninehills/skills, 280 stars), Read URLs and PDFs (tw93/Waza, 7.2k stars), Markitdown (ImCa0/just-laws, 781 stars) and Gzh Design (isjiamu/gzh-design-skill, 4k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Financial Expense Automation?

DjangoPeng (a GitHub user) maintains it in DjangoPeng/agentic-ai, which has 152 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on July 8, 2026.

Source: DjangoPeng/agentic-ai on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.