Skill

extract-invoices

Use when the user wants to extract invoice information from emails, process PDF attachments for invoice data, pull out amounts and IBANs from invoices, or parse invoice PDFs.

Install

npx claudepluginhub karimstekelenburg/k3m-marketplace --plugin personal-finance

Tool Access

This skill is limited to using the following tools:

Bash

SKILL.md

Similar Skills

cache-components

Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.

cache-components

138.8k

claude-opus-4-5-migration

2 files

Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.

claude-opus-4-5-migration

83.2k

bmad-help

Analyzes BMad project state from catalog CSV, configs, artifacts, and query to recommend next skills or answer questions. Useful for help requests, 'what next', or starting BMad.

bmad-pro-skills

43.8k

Stats

Parent Repo Stars0

Parent Repo Forks0

Last CommitMar 29, 2026

Actions

View Source View Plugin View on GitHub View README

Parse arguments from $ARGUMENTS. Determine the data directory: use --data-dir if provided, otherwise default to ~/Documents/finance-data/.

Determine the extraction scope from the arguments:

If --email-id ID is given: Run the extractor for that single email's attachment:

uv run ${CLAUDE_PLUGIN_ROOT}/scripts/extract_invoice.py --email-id ID [--data-dir PATH]

If --all-pending is given (or no scope flag is provided): Run the extractor over all emails with status='pending' in the database:

uv run ${CLAUDE_PLUGIN_ROOT}/scripts/extract_invoice.py --all-pending [--data-dir PATH]

After the script completes, report:

Total invoices processed
How many succeeded with high confidence (no fallback)
How many used llm_fallback (flag these clearly — they require review)

Display a summary table for all extracted invoices:

Email ID	Vendor	Amount	Due Date	Confidence	LLM Fallback

Mark llm_fallback rows with a warning indicator.

For each llm_fallback invoice, perform manual extraction:

Show the raw extracted text from the script's output for that invoice
Read the text carefully and extract: vendor name, invoice number, total amount (with currency), IBAN, and due date
Present your extracted values to the user for confirmation

Once confirmed (or if the values are unambiguous), write the corrected fields to the database by running:

uv run ${CLAUDE_PLUGIN_ROOT}/scripts/extract_invoice.py --update-record --email-id ID --vendor "..." --amount ... --due-date ... --iban ... [--data-dir PATH]

Mark the record's llm_fallback flag as resolved after writing

If any attachment cannot be read (corrupt PDF, unsupported format), report the email ID and error; continue processing remaining items.