[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"pack-detail-finance-accountant-ai-pack-es":3,"seo:pack:finance-accountant-ai-pack:es":98},{"code":4,"message":5,"data":6},200,"操作成功",{"pack":7},{"slug":8,"icon":9,"tone":10,"status":11,"status_label":12,"title":13,"description":14,"items":15,"install_cmd":97},"finance-accountant-ai-pack","💰","#16A34A","new","Nuevo · esta semana","Pack de IA para Finanzas y Contabilidad","Diez picks para el contable, analista FP&A o CFO de pyme que pone IA sobre los libros: OCR de recibos, automatización de hojas, libros en texto plano, facturación, agentes de modelado y dashboards de reporte. Construido alrededor de una pista auditable, no de un chatbot.",[16,28,35,43,53,61,68,75,83,90],{"id":17,"uuid":18,"slug":19,"title":20,"description":21,"author_name":22,"view_count":23,"vote_count":24,"lang_type":25,"type":26,"type_label":27},758,"3ac555d9-d75c-4208-ba46-974e4a717234","zerox-zero-shot-pdf-ocr-ai-pipelines-3ac555d9","Zerox — Zero-Shot PDF OCR for AI Pipelines","Extract text from any PDF using vision models as OCR. Zerox converts PDF pages to images then uses GPT-4o or Claude to extract clean markdown without training.","Script Depot",205,0,"en","skill","Skill",{"id":29,"uuid":30,"slug":31,"title":32,"description":33,"author_name":22,"view_count":34,"vote_count":24,"lang_type":25,"type":26,"type_label":27},210,"42976daf-a56a-4152-9afb-d5b00d130a08","marker-convert-pdf-markdown-high-accuracy-42976daf","Marker — Convert PDF to Markdown with High Accuracy","Fast, accurate PDF to Markdown + JSON converter. Handles tables, images, equations, code blocks, and multi-column layouts. GPU-accelerated. 33K+ GitHub stars.",139,{"id":36,"uuid":37,"slug":38,"title":39,"description":40,"author_name":41,"view_count":42,"vote_count":24,"lang_type":25,"type":26,"type_label":27},68,"b9536815-4d1b-449f-ab72-bee80e34a3ed","claude-official-skill-xlsx-spreadsheet-operations-b9536815","Claude Official Skill: XLSX — Spreadsheet Operations","Claude Code skill for Excel spreadsheets. Create workbooks, write formulas, generate charts, pivot tables, and data analysis.","Anthropic",147,{"id":44,"uuid":45,"slug":46,"title":47,"description":48,"author_name":49,"view_count":50,"vote_count":24,"lang_type":25,"type":51,"type_label":52},3689,"386c6339-7093-59bf-afe3-4c9abf849240","excel-mcp-server-edit-xlsx-without-excel","Excel MCP Server — Edit .xlsx Without Excel","excel-mcp-server is an MCP server to edit .xlsx without Microsoft Excel; verified 3,820★ and runs via `uvx excel-mcp-server` (stdio or streamable HTTP).","MCP Hub",69,"mcp","MCP",{"id":54,"uuid":55,"slug":56,"title":57,"description":58,"author_name":59,"view_count":60,"vote_count":24,"lang_type":25,"type":26,"type_label":27},2187,"070cdf4e-418d-11f1-9bc6-00163e2b0d79","taxhacker-self-hosted-ai-accounting-receipts-invoices-070cdf4e","TaxHacker — Self-Hosted AI Accounting for Receipts and Invoices","A self-hosted AI-powered accounting app that scans receipts and invoices, categorizes expenses with LLM analysis, and tracks finances with custom prompts and multi-currency support.","AI Open Source",94,{"id":62,"uuid":63,"slug":64,"title":65,"description":66,"author_name":59,"view_count":67,"vote_count":24,"lang_type":25,"type":26,"type_label":27},4589,"2468a83a-5426-11f1-9bc6-00163e2b0d79","hledger-robust-plain-text-accounting-tool-2468a83a","Hledger — Robust Plain-Text Accounting Tool","A cross-platform accounting program for tracking money, time, or any quantity using plain-text journal files, with terminal, web, and API interfaces.",44,{"id":69,"uuid":70,"slug":71,"title":72,"description":73,"author_name":22,"view_count":74,"vote_count":24,"lang_type":25,"type":26,"type_label":27},1749,"6887eac2-3b87-11f1-9bc6-00163e2b0d79","invoice-ninja-open-source-invoicing-payments-6887eac2","Invoice Ninja — Open Source Invoicing and Payments","Invoice Ninja is a free, open-source platform for invoicing, payments, expenses, and time tracking. It supports over 45 payment gateways and can be self-hosted or used as a managed service.",112,{"id":76,"uuid":77,"slug":78,"title":79,"description":80,"author_name":81,"view_count":82,"vote_count":24,"lang_type":25,"type":26,"type_label":27},4338,"94144188-39a7-4495-b4a1-562536ab0242","claude-code-agent-quant-analyst-94144188","Claude Code Agent: Quant Analyst","Quantitative finance and algorithmic trading specialist. Use PROACTIVELY for financial modeling, trading strategy development, backtesting, risk analysis, and portfolio...","TokRepo精选",28,{"id":84,"uuid":85,"slug":86,"title":87,"description":88,"author_name":81,"view_count":89,"vote_count":24,"lang_type":25,"type":26,"type_label":27},4360,"2d86ab81-cf4e-49d9-a55c-6fe84842d8ed","claude-code-agent-report-generator-2d86ab81","Claude Code Agent: Report Generator","Use this agent when you need to transform synthesized research findings into a comprehensive, well-structured final report. This agent excels at creating readable narratives...",38,{"id":91,"uuid":92,"slug":93,"title":94,"description":95,"author_name":22,"view_count":96,"vote_count":24,"lang_type":25,"type":26,"type_label":27},2112,"61a7b500-40a1-11f1-9bc6-00163e2b0d79","dash-data-apps-dashboards-pure-python-61a7b500","Dash — Data Apps and Dashboards in Pure Python","Dash by Plotly lets you build interactive analytical web applications entirely in Python, with no JavaScript required. It powers data dashboards across finance, biotech, and engineering.",128,"tokrepo install pack\u002Ffinance-accountant-ai-pack",{"pageType":99,"pageKey":8,"locale":25,"title":100,"metaDescription":101,"h1":102,"tldr":103,"bodyMarkdown":104,"faq":105,"schema":121,"internalLinks":127,"citations":140,"wordCount":153,"generatedAt":154},"pack","Finance + Accountant AI Pack — 10 Tools to Close the Books With AI","Zerox, Marker, Claude XLSX, Excel MCP, TaxHacker, Hledger, Invoice Ninja, Quant Analyst agent, Report Generator, Dash — the 10-asset stack an accountant or FP&A analyst uses to OCR receipts, categorize, model, and report without losing the audit trail.","Finance + Accountant AI Pack — From Receipt to Reporting","Ten picks in deliberate order: get paper into structured data first (OCR + parsing), put it into a sheet or ledger you can audit, let an agent do the modeling and forecasting, then ship a dashboard the CFO actually opens. Built around an audit trail, not a chatbot.","## What's in this pack\n\nThis is the stack for someone who actually has to **close the month** — a small-business accountant, a one-person FP&A team at a Series A startup, a CFO who still has to touch the books between board meetings. Not a Twitter list of 30 AI finance startups. Every pick here does one specific job in the close cycle: turn paper into data, get it into something auditable, categorize and analyze, model the future, and report it.\n\nThe whole pack is built around one principle: **AI accelerates the work, but the audit trail is non-negotiable**. That's why the ledger pick is plain-text (`hledger`), not a black-box SaaS. That's why categorization runs through a self-hosted app you can inspect (TaxHacker), not someone else's API. And it's why the spreadsheet picks let an agent *edit your existing workbook* instead of replacing it.\n\n## Install in this order (ingestion → categorization → analysis → forecasting → reporting)\n\n1. **Zerox — Zero-Shot PDF OCR for AI Pipelines** — start here. Receipts, vendor invoices, and bank statements arrive as PDFs and JPGs. Zerox uses vision models to extract structured data in one shot, no template training. This is the front door.\n2. **Marker — Convert PDF to Markdown with High Accuracy** — Zerox is great for single receipts; Marker shines on the long stuff (multi-page statements, 10-Ks, audit reports). Markdown out, tables preserved. Pair them.\n3. **Claude Official Skill: XLSX — Spreadsheet Operations** — once your data is structured, you need a workbook. The official Claude skill reads\u002Fwrites `.xlsx` cleanly, including formulas and formatting. This is the bridge between \"raw extract\" and \"thing you can show your boss.\"\n4. **Excel MCP Server — Edit .xlsx Without Excel** — MCP exposure of the same capability to your local agent. Lets Claude Code or Cursor open your monthly close workbook, append rows, recompute totals, and save — without you ever leaving the chat. Critical for repetitive period-end work.\n5. **TaxHacker — Self-Hosted AI Accounting for Receipts and Invoices** — self-hosted app that scans receipts, categorizes expenses with an LLM, and tracks finances with multi-currency. Custom prompts mean you can tune it to your chart of accounts. The categorization layer that doesn't ship your data to a vendor.\n6. **Hledger — Robust Plain-Text Accounting Tool** — the audit backbone. Every categorized transaction lands in a plain-text journal you can `git diff`, review, and roll back. Double-entry, reconciliation, balance sheets, P&L — all from one file you actually own.\n7. **Invoice Ninja — Open Source Invoicing and Payments** — the AR side. Send invoices, track payments, reconcile receivables back into `hledger`. Self-hosted so client data stays on your server.\n8. **Claude Code Agent: Quant Analyst** — financial modeling specialist subagent. Use it for valuation, scenario analysis, sensitivity tables, and DCFs. Once your data is clean in XLSX\u002F`hledger`, this is the agent you point at it to do the FP&A heavy lifting.\n9. **Claude Code Agent: Report Generator** — monthly board deck, variance commentary, narrative around the numbers. The Quant Analyst gives you the model; the Report Generator gives you the slide.\n10. **Dash — Data Apps and Dashboards in Pure Python** — for the always-on view. Build a P&L dashboard, cash runway tracker, or AR aging report once in pure Python; refresh it on a schedule. Replaces \"send me the latest sheet\" forever.\n\n## How they fit together\n\n```\n   Paper \u002F PDF \u002F JPG\n         │\n   ┌─────┴──────┐\n   ▼            ▼\n Zerox       Marker\n (1-page)  (multi-page,\n            statements)\n   │            │\n   └─────┬──────┘\n         ▼\n  structured rows \u002F markdown\n         │\n   ┌─────┴──────────┐\n   ▼                ▼\n Claude XLSX   TaxHacker\n (workbook)    (categorize\n   │            + GL coding)\n Excel MCP          │\n (agent edits)      ▼\n   │            Hledger\n   └─────┬──────(audit trail)\n         ▼              ▲\n   Invoice Ninja ───────┘\n   (AR side feeds back)\n         │\n         ▼\n   ┌──── Quant Analyst ────┐\n   │  (modeling, sensitivity)│\n   └────┬───────────────────┘\n        ▼\n   Report Generator      Dash\n   (board deck text)  (live dashboard)\n```\n\nThe critical join is **Hledger + the agents**: as long as the source of truth is one plain-text file, you can let agents read it, model on top, and write reports — without ever giving them write access to your books. Read-only AI on auditable data is the safe shape.\n\n## Tradeoffs you'll hit\n\n- **AI categorization vs rule-based** — TaxHacker (or any LLM categorizer) will guess your chart of accounts brilliantly 90% of the time and quietly miscode 10%. Rule-based (`hledger`'s built-in `auto`-postings, regex rules) is boring but reproducible. Best practice: rules first for the predictable 80%, LLM only for the long tail. Always review.\n- **Monthly close vs continuous close** — automation tempts you toward \"close every day.\" Resist for the first quarter. Run your first three closes on a monthly cadence so you learn where the categorization fails. *Then* go continuous, with anomaly checks instead of human review.\n- **Self-hosted vs SaaS for sensitive data** — every tool in this pack except the Claude agents can be self-hosted. That's deliberate. Vendor invoices and payroll data are exactly the categories you don't want flowing to a SaaS you can't audit. Pay the 1-2 hours of Docker setup; save yourself the SOC 2 conversation later.\n- **Plain-text accounting vs QuickBooks\u002FXero** — `hledger` is uncompromising and ugly. QuickBooks is friendly and a walled garden. If your business is \u003C20 transactions\u002Fweek and you can read text, `hledger` wins on auditability. If you have employees who need to enter invoices, stay on QuickBooks and let the AI agents read its exports — don't make staff learn Ledger syntax.\n\n## Common pitfalls\n\n- **AI hallucinating numbers** — the single most dangerous failure mode in finance AI. An LLM that summarizes a P&L will *invent* line items that look plausible. Mitigation: never let the agent produce the number; let it produce the *query* (SQL, hledger command, formula) and run it deterministically. Then explain what it returns.\n- **Lack of audit trail** — if you can't answer \"why did this entry hit this account on this date\" three months later, the audit will be miserable. Commit your `hledger` journal to git, tag each monthly close, and keep the OCR source PDFs in the same repo. Boring infrastructure that saves your year-end.\n- **Letting an agent write to the GL directly** — don't. Agents propose journal entries to a staging file; a human (you) reviews and commits to the main journal. The latency cost is 5 minutes\u002Fweek; the cost of an agent silently doubling an accrual is six hours of forensic work.\n- **OCR confidence without verification** — Zerox will happily return $1,200.00 when the receipt says $12.00. Always reconcile the OCR'd total against the credit card statement before posting. Two-way match is not optional.\n- **Forgetting the legal layer** — this pack is editorial guidance for a workflow. It does not replace your CPA, your tax advisor, or jurisdiction-specific bookkeeping rules. The tools make the work faster; they don't make you compliant. Talk to a human professional before any tax-relevant decision.",[106,109,112,115,118],{"q":107,"a":108},"Is any of this safe to use for a real business's books, or is it just toys?","It's real, with caveats. Hledger has been used for company books since 2007; Invoice Ninja runs production billing for many small businesses; Zerox and Marker are production OCR. The risk surface is the AI layer — TaxHacker, the Claude agents, the spreadsheet automation. Treat those as junior staff: every output gets reviewed before it hits the ledger. Used that way, the pack is safe for SMB books. It is not yet appropriate for regulated industries (banks, insurance) without a controls overlay your auditor signs off on.",{"q":110,"a":111},"What's the smallest version of this pack I can start with this week?","Three picks: Zerox (OCR receipts into structured rows), Claude XLSX skill (paste those rows into your existing monthly workbook), and Hledger (commit the categorized output to a plain-text journal). That's a working close pipeline in one afternoon. Add TaxHacker when manual categorization becomes the bottleneck. Add the Quant Analyst agent the first time you have to build a 3-statement model. Don't install everything at once — the categorization step needs to be tuned against *your* chart of accounts before later tools matter.",{"q":113,"a":114},"Why plain-text accounting (Hledger) instead of QuickBooks or Xero?","Three reasons: auditability (every change is a git diff), AI-friendliness (an agent can read a 10MB text journal in one prompt — try that with a QuickBooks backup), and ownership (your books don't disappear if a vendor raises prices or shuts down). The tradeoff is staff onboarding: nobody other than a finance-literate engineer wants to type Ledger syntax. If you have a bookkeeper who lives in QuickBooks, keep QuickBooks as the data-entry layer and pipe its exports into hledger for AI analysis. Hybrid is the realistic answer.",{"q":116,"a":117},"Can the Claude Code agents actually do FP&A work, or is this hype?","They're useful, not magic. The Quant Analyst agent is genuinely good at: scenario tables, sensitivity analysis, ratio computation against a clean dataset, model documentation, and explaining a model someone else built. It is mediocre at: judgment calls (what's a reasonable churn assumption), industry-specific benchmarks (it doesn't know your competitive set), and anything requiring access to private data unless you wire it in. Use it as a senior associate, not a partner. The human signs the model.",{"q":119,"a":120},"How do I keep an audit trail when AI is touching the books?","Three rules. First, agents propose; humans commit. Every AI-generated entry lands in a staging file (`pending.journal`), gets a human review pass, then merges into the main journal with a git commit. Second, source documents stay with the data. PDFs of receipts and invoices live in the same git repo as the journal, named by transaction ID. Three months from now you can trace any entry to the original document in two clicks. Third, version every monthly close as a git tag (e.g. `close-2026-04`). If something needs to be restated, you have an immutable record of what the books said when you closed. None of this is more work than a normal accountant already does; it's the same controls in a git-shaped wrapper.",{"@context":122,"@type":123,"name":124,"description":125,"numberOfItems":126,"inLanguage":25},"https:\u002F\u002Fschema.org","ItemList","Finance + Accountant AI Pack","Ten AI-era picks for accountants, FP&A analysts, and small-business CFOs: receipt OCR, spreadsheet automation, plain-text accounting, invoicing, modeling agents, and reporting dashboards.",10,[128,132,136],{"url":129,"anchor":130,"reason":131},"\u002Fen\u002Fai-tools-for\u002Fdata-analysis","AI tools for data analysis","FP&A and accounting workflows live inside the broader data-analysis catalog on TokRepo",{"url":133,"anchor":134,"reason":135},"\u002Fen\u002Fai-tools-for\u002Fautomation","Automation tools for AI agents","Excel MCP and agent-driven close steps belong to the wider automation catalog",{"url":137,"anchor":138,"reason":139},"\u002Fen\u002Ffeatured","Featured assets on TokRepo","These ten picks are part of the broader curated catalog of agent-ready assets",[141,145,149],{"claim":142,"source_name":143,"source_url":144},"Hledger is a robust plain-text double-entry accounting tool","hledger official site","https:\u002F\u002Fhledger.org\u002F",{"claim":146,"source_name":147,"source_url":148},"Zerox provides zero-shot OCR for PDFs aimed at AI pipelines","Zerox on GitHub","https:\u002F\u002Fgithub.com\u002Fgetomni-ai\u002Fzerox",{"claim":150,"source_name":151,"source_url":152},"Marker converts PDFs to Markdown with table preservation","Marker on GitHub","https:\u002F\u002Fgithub.com\u002FVikParuchuri\u002Fmarker",1140,"2026-05-22T12:00:00Z"]