Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add panaversity/agentfactory-business-plugins --skill bank-reconciliationgit clone --depth 1 https://github.com/panaversity/agentfactory-business-pluginsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/panaversity/agentfactory-business-plugins/bank-reconciliation)<a href="https://agentmods.dev/skills/panaversity/agentfactory-business-plugins/bank-reconciliation"><img src="https://agentmods.dev/badge/skills/panaversity/agentfactory-business-plugins/bank-reconciliation/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/panaversity/agentfactory-business-plugins/bank-reconciliation"><img src="https://agentmods.dev/badge/skills/panaversity/agentfactory-business-plugins/bank-reconciliation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00120 | $0.02340 |
| Opus 5 | $0.00060 | $0.01170 |
| Sonnet 5 | $0.00024 | $0.00468 |
| Haiku 4.5 | $0.00012 | $0.00234 |
Grade A, and why
bank-reconciliation scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 232 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RECONCILIATION TYPES — LOAD THE RIGHT CONTEXT
| Task Description | Use This Context |
|---|---|
| Nostro / correspondent account | sections: NOSTRO, AGEING SLA, MATCHING HIERARCHY |
| IFRS 9 provision GL tie-out | sections: FOUR-WAY RECON, P&L TIE-OUT |
| Suspense account clearance | sections: SUSPENSE, AGEING SLA, WRITE-OFF |
| Inter-company / inter-branch | sections: INTERCO, AGEING SLA |
| Trade / securities settlement | sections: TRADE SETTLEMENT, AGEING SLA |
| Full financial close | ALL sections |
MATCHING HIERARCHY — APPLY IN THIS EXACT ORDER
Step 1 — EXACT MATCH Criteria: amount = amount AND value date = value date AND reference = reference Action: Auto-clear. No human review required. Log as matched.
Step 2 — FUZZY MATCH Criteria: amount = amount AND value date = value date AND reference partially matches (truncated, reformatted, leading zeros stripped, SWIFT truncation applied) Action: Auto-clear with exception note. Flag for daily exception report review.
Step 3 — DATE TOLERANCE MATCH Criteria: amount = amount AND reference matches AND value date within ±2 business days Action: Auto-clear with timing difference notation. Post value date correction if needed.
Step 4 — PARTIAL SUM POOL Criteria: Multiple items on one side aggregate to a single item on the other side (common in retail settlement aggregation, netting arrangements) Action: Group items into pool. Present to human analyst for confirmation before clearing.
Step 5 — UNMATCHED RESIDUAL Criteria: No rule above is satisfied Action: Generate investigation hypothesis (see BREAK INVESTIGATION below). Queue for analyst with hypothesis, probability estimate, and recommended action.
BREAK CLASSIFICATION
Mirror item only (no external statement item): → TIMING DIFFERENCE: Payment sent / instructed but not yet appearing on statement Probability: High for items ≤ 2 business days Action: Monitor for settlement; query counterparty if aged > 3 days → PAYMENT FAILURE / REJECTION: Payment rejected by correspondent, no credit applied Probability: Rises sharply after 2 days with no statement match Action: Check SWIFT gpi tracker; request SWIFT MT199 status query
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 232 lines · 120 tokens per session scan A 1e4fff46e9b8
bank-reconciliation is a skill published in the GitHub repository panaversity/agentfactory-business-plugins (29 stars, last pushed 5mo ago), licensed Apache-2.0. It adds 120 tokens to every session and 2,340 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
roi-estimator
Use when estimating the ROI of adopting an LLM/agent for a workflow, with structured assumptions.
mcp_paypal
Operate PayPal through its official remote MCP in the configured sandbox or live environment. Use when the user asks to create/send invoices, manage orders and payments, issue refunds, handle disputes, manage products or subscriptions, track shipments, or read transactions.
mcp_stripe
Operate Stripe through the official remote Stripe MCP (mcp.stripe.com). Use when the user asks to read or update Stripe API resources, issue refunds, plan an integration, run reports, inspect the connected account, or search Stripe documentation.
mcp_quickbooks
Operate the user's QuickBooks Online company through the QuickBooks MCP. Use when the user asks to look up or create customers and invoices, send or void an invoice, review payments/bills/accounts/vendors, or run a financial report.
mcp_square
Operate the user's Square account through the Square MCP. Use when the user asks to look up locations, catalog items, orders, or customers, create catalog items or customers, or read/adjust inventory counts on Square.
finance-ledger
Read, query, and record validated income, expense, invoice, bill, payment, refund, transfer, adjustment, and journal-entry facts in the active Workspace Finance Ledger.