Amortization schedule > loan servicing payment waterfall recompute > RESPA aggregate escrow analysis. Three independent recomputations over one loan's servicing math. Each step computes from its own declared inputs; the order is analytical, not a statutory sequence.
Copy this paragraph into Claude, OpenClaw, or any MCP-aware agent to run this whole workflow on the server and check one field that answers a real question.
Question: The servicer applied this month's payment and says escrow is fine. Does an independent recompute of the split and the aggregate escrow position agree before we post the cycle?
Run the AINumbers MCP tool `run_chain` with {"chain":"servicing-audit"} and read `art-664-loan-servicing-waterfall-recompute.verdict`.
Re-run with `inputs` for `art-664-loan-servicing-waterfall-recompute`, reusing the values that result echoed, with `payment_amount` set to 9000. Compare the same field.
Verify: call `verify_execution_hash` (https://mcp.ainumbers.co/mcp) with `claimed_hash` set to `composite_execution_hash` and the full `composite_artifact`.
Ledger, for a human re-check: https://ledger.ainumbers.co/
PII rule: send synthetic or anonymised inputs only. The MCP server runs these kernels and logs no payloads.
Chain page: https://ainumbers.co/chaingraph/chains/servicing-audit.html