AP Invoice Processing
Automates the clerical front end of accounts payable while keeping the approve-and-pay decision with a person. It scans an AP inbox for vendor invoices, reads attachments and scans, and extracts invoice date, vendor, description, invoice number, amount, and due date, citing where each value came from. It dedupes against your AP log by invoice number so nothing is logged twice, appends one new row per invoice marked 'added by skill, needs review', forwards the email to your AP system (for example bill.com or Ramp — an example, not a requirement), applies a label, archives the email, and hands back a run summary of logged / skipped / uncertain. Vendors on your exclude list are skipped so card-paid charges handled by the sibling card-receipt-forwarding flow are not double-processed. The safety guardrail is the point: the skill never approves and never pays. It drafts and files; a human audits every new row before any money moves, and anything unreadable or conflicting is flagged 'uncertain, needs review' rather than guessed. Treat it like a sharp intern — about ninety percent right, fast on the boring parts, and always audited. Detection criteria, the exclude list, and the scheduled-run lookback window are operator settings you confirm in plan mode before the first run.
Simple install — no setup required
Download the skill file below
Add it to Claude — pick your platform:
Claude desktop app
- Open the Claude desktop app
- Go to Code, then Customize
- Click Create a new skill
- Upload the skill file you downloaded
claude.ai (web)
- Go to Customize, then Skills
- Click +, then Create a new skill
- Upload the skill file you downloaded
Say a trigger phrase to activate it
Download the skill file below
Place it at .cursor/skills/ap-invoice-processing/SKILL.md (rename the downloaded file to SKILL.md). The shared .agents/skills/ap-invoice-processing/SKILL.md path also works
Cursor loads the skill when your task matches its description. Say a trigger phrase to activate it
Download the skill file below
In ChatGPT, create a Project and attach the file (or paste its contents into the project instructions). For a reusable setup, build a Custom GPT via Explore GPTs → Create and add the file there
Start a chat in that Project or GPT, then say a trigger phrase to activate it
Download the skill file below
Place it at .github/skills/ap-invoice-processing/SKILL.md in your repository (rename the downloaded file to SKILL.md)
Copilot loads the skill when it is relevant to your task. Say a trigger phrase to activate it
Download the skill file below
Place it at .agents/skills/ap-invoice-processing/SKILL.md in your repository (rename the downloaded file to SKILL.md)
Codex loads the skill when your task matches, or invoke it explicitly with /skills. Say a trigger phrase to activate it
Your skill is downloading.
Get an email when we ship new skills and updates.
You're on the list. We'll email you when new skills drop.
What this skill does
This is an overview of the skill from the site, not the skill's real instructions. The full SKILL.md Claude reads when this skill is active lives on GitHub at the pinned source below.
AP Invoice Processing automates the clerical front end of accounts payable and leaves the approve-and-pay decision with a person. It watches an AP inbox for vendor invoices, reads the attachments and scans, and pulls out invoice date, vendor, description, invoice number, amount, and due date — citing where each value came from.
Each invoice is deduped against your AP log, appended as one new row marked “added by skill, needs review,” forwarded to your AP system, labeled, and archived. Every run hands back a summary of what was logged, skipped, and flagged uncertain. Vendors on your exclude list are skipped so charges already handled by the sibling card-receipt-forwarding flow are not double-processed.
The guardrail is the point: it never approves and never pays. It drafts and files; a human audits every new row before any money moves, and anything unreadable or conflicting is flagged “uncertain, needs review” rather than guessed. Detection criteria, the exclude list, and the scheduled-run lookback window are operator settings you confirm before the first run.
View full SKILL.md on GitHub