- Home
- Skills
- Finance & Accounting
- Invoice CSV Reconciliation & Exception Finder
More screenshots
Works with the AI Agents you already use
Invoice CSV Reconciliation & Exception Finder
Reconcile invoice, payment and ledger CSVs locally to flag missing matches, amount differences, duplicates and unresolved exceptions before close.
$7
Invoice CSV Reconciliation & Exception Finder
Example session with this skill installed
Run the included sample invoice and payment CSV files through the local reconciliation pack using its sample policy. Show deterministic matches and exceptions. Do not post an accounting entry or access a bank feed.
- Read your context and instructions
- Compiled the invoice csv reconciliation
- Generated the data export
Invoice Payment CSV Reconciliation — included sample
Status
REVIEW. The sample contains 2 invoices and 2 payments. One pair matches deterministically and two items require review.
| Invoice | Payment | Method | Amount | Result |
|---|---|---|---|---|
| INV-100 | PAY-1 | explicit_invoice_id | USD 125.00 | Matched; difference USD 0.00 |
| INV-101 | — | — | — | UNMATCHED_INVOICE |
| — | PAY-2 | — | — | UNMATCHED_PAYMENT |
The packet includes result.json, matches.csv, exceptions.jsonl, duplicate-candidates.csv, RECONCILIATION.md and evidence-receipt.json. The included policy requires currency match, allows a 45-day date window and a 10-minor-unit amount tolerance. Review INV-101 and PAY-2 against the source exports before any bookkeeping decision. Matching uses only supplied IDs, references, amounts, customer IDs, dates and policy; no accounting entry is posted. No OCR, live bank feed, split payment or currency conversion is included.
matches.csv
CSV · data export
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Invoice and payment exports often use different identifiers, dates and amount formats, leaving close teams to investigate mismatches by hand. This local tool normalizes supplied CSV files, matches supported records and produces separate matched, exception and review outputs with a reproducible summary. It does not connect to a bank or accounting system, post entries or provide accounting advice.
This downloadable local package turns two authorized buyer-supplied CSV exports into a fixed, reproducible reconciliation review packet. It includes an installable Python CLI, a local stdio MCP adapter and a companion Agent Skill. One run creates result.json, matches.csv, exceptions.jsonl, duplicate-candidates.csv, RECONCILIATION.md and evidence-receipt.json. The core uses supplied IDs, references, amounts, customer IDs, dates and policy for one-to-one matching. It rejects unsafe paths and malformed inputs, never overwrites an existing packet, escapes formula-active CSV cells and binds each input and output with SHA-256. Processing stays local; no API key, login, upload, telemetry or live external action is included. Useful for finance operations teams, bookkeepers, small businesses and agencies preparing reviewable handoff evidence before a risky action. Requires Windows 11, Python 3.11–3.14 and local evidence files in the documented schemas; optional local MCP adapter uses mcp==2.2.0. Limitations: no OCR, bank feed, accounting write, tax decision, split payment or currency conversion. A human must review exceptions and decide any accounting action.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Finance & Accounting.
No OCR, live bank feed, accounting write, tax decision, split-payment allocation or currency conversion. Inputs must be authorized local CSV files in documented schemas. A human reviews exceptions before any bookkeeping action.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 2 months ago
- Software
Python
Permissions
Runs a local Python CLI or optional stdio MCP adapter; reads two buyer-selected CSV exports and a policy file; writes a new reconciliation packet. It does not connect to banks or accounting systems. No browser, network or environment-variable access is required.