Works with the AI Agents you already use
ap three way match
Deterministic three-way reconciliation for AP invoices, POs, and receipts with cryptographic audit trails.
$19.99
/moSubscription · cancel anytime
Secure checkout via Stripe
- Always the latest version
- Delivered live to your agent via MCP
- Cancel anytime, access ends at period end
ap three way match
Example session with this skill installed
Reconcile PO #8821: Unit price $45.00, Qty 100. Received 95 units. Invoice shows $47.25 per unit for 95 units. Price tolerance is 2%.
- Read your context and instructions
- Compiled the ap three way
I have performed the three-way reconciliation for PO #8821. The system calculated a price variance of 5.0%, which exceeds your 2.0% threshold, resulting in a price mismatch status.
Reconciliation Report: PO #8821
| Metric | Value |
|---|---|
| PO Total (Expected) | $4,500.00 |
| Invoiced Total (Actual) | $4,488.75 |
| Price Variance % | 5.0% |
| Quantity Variance | 0 units |
| Reconciliation Status | PRICE_MISMATCH |
| Audit Hash | f7b1291a0c2d489e812f9432d0a87123ef98201b4bc893d102e3b12984bc2104 |
Detailed Breakdown
- The invoiced unit price of $47.25 represents a 5% increase over the PO price of $45.00.
- While the invoiced quantity (95) matches the received quantity (95), the price deviation exceeds the allowed 2% tolerance.
- Payment is currently flagged and not recommended for automatic approval.
Next steps
- Contact the vendor to dispute the $2.25 unit price discrepancy.
- Update the Purchase Order if the price increase was pre-approved but not documented.
- Re-run reconciliation once the corrected invoice is received.
Subscription · always the latest version
- Secure checkout via Stripe
- Cancel anytime
- Delivered live via MCP
About this skill
Manual invoice reconciliation is prone to human error and expensive overhead. This skill automates line-item three-way matching by programmatically comparing Purchase Orders, Receiving Slips, and Vendor Invoices against strict price and quantity tolerances.
What it does
- Line-item matching performs granular comparisons between PO, receipt, and invoice data points.
- Variance calculation computes price deviations as percentages and quantity discrepancies in units.
- Tolerance enforcement applies user-defined percentage thresholds to flag price mismatches automatically.
- Automated reconciliation assigns statuses like MATCHED_APPROVED, PRICE_MISMATCH, or OVERBILLING_DISCREPANCY based on deterministic logic.
- Integrity sealing generates a SHA-256 audit hash for every successful reconciliation trace to ensure data consistency.
How it works
- Input validation checks caller parameters against schema constraints for price, quantity, and tolerance levels.
- Deterministic execution runs the core matching formulas to calculate totals and variances.
- Status assignment evaluates the calculated variances against the provided tolerance thresholds.
- Audit generation produces a final JSON response containing the arithmetic breakdown and a verification hash.
Frameworks & tools
The skill operates via an MCP-compliant JSON-RPC interface and uses SHA-256 hashing for audit trails. It is designed for integration with enterprise FinOps stacks and ERP systems.
Why this beats prompting it yourself
Standard LLM prompting often struggles with precise floating-point arithmetic and consistent enforcement of mathematical bounds. This skill uses a deterministic execution engine to ensure calculations are mathematically sound and repeatable every time.
Use cases
- AP Automation to process high volumes of vendor invoices without manual intervention.
- Audit Logging to maintain a verifiable record of reconciliation decisions with cryptographic hashes.
- Discrepancy Detection to flag overbilling or unit price creep before payment processing.
Known limitations
Requires numeric inputs within strict mathematical bounds. It does not provide professional financial advice or replace final management approval for payments.
Input must adhere to schema.json constraints. Limited to quantitative reconciliation; does not perform OCR or document parsing.
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 any payment.
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
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 16 days ago
Permissions
No special permissions declared or detected