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 peerjakobsen/smartspender --skill transaction-matchinggit clone --depth 1 https://github.com/peerjakobsen/smartspenderWrote 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/peerjakobsen/smartspender/transaction-matching)<a href="https://agentmods.dev/skills/peerjakobsen/smartspender/transaction-matching"><img src="https://agentmods.dev/badge/skills/peerjakobsen/smartspender/transaction-matching/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/peerjakobsen/smartspender/transaction-matching"><img src="https://agentmods.dev/badge/skills/peerjakobsen/smartspender/transaction-matching.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 | $0.00024 | $0.03236 |
| Opus 5 | $0.00012 | $0.01618 |
| Sonnet 5 | $0.00005 | $0.00647 |
| Haiku 4.5 | $0.00002 | $0.00324 |
Grade A, and why
transaction-matching 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 9d 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 — 326 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Transaction Matching
Purpose
Provides rules for linking an uploaded receipt to an existing bank transaction in transactions.csv. Uses amount, date, and merchant name as matching signals with confidence scoring.
Matching Signals
Three signals determine whether a receipt matches a transaction:
| Signal | Weight | Rule |
|---|---|---|
| Amount | Primary | Receipt total (positive) must match transaction amount (negative) within +-1% |
| Date | Primary | Receipt date must be within +-1 day of transaction date |
| Merchant | Secondary | Normalized merchant name should match (boosts confidence but not required) |
Both amount and date must match for a candidate. Merchant match increases confidence but a missing merchant match does not disqualify.
Amount Comparison
Receipts store totals as positive numbers. Transactions store expenses as negative numbers. When comparing:
- Take the absolute value of the transaction amount
- Compare with the receipt total
- Allow +-1% tolerance
Formula: |receipt_total - |transaction_amount|| / receipt_total <= 0.01
Why +-1%: Small rounding differences between what the store charges and what the bank posts (e.g., currency conversion micro-adjustments, tip rounding).
Examples
| Receipt Total | Transaction Amount | Difference | Match? |
|---|---|---|---|
| 347.50 | -347.50 | 0.00% | Yes |
| 347.50 | -348.00 | 0.14% | Yes |
| 347.50 | -344.00 | 1.01% | No |
| 299.00 | -299.00 | 0.00% | Yes |
Date Window
Card transactions often post 1 business day after the purchase. Allow a +-1 day window:
- Receipt date: 2026-01-28
- Valid transaction dates: 2026-01-27, 2026-01-28, 2026-01-29
Why +-1 day: Dankort and Visa purchases at end of day may post the next morning. Some merchants batch transactions overnight.
Merchant Name Matching
Compare the receipt's normalized merchant name against the transaction's description using the same normalization rules from skills/categorization/SKILL.md:
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.
- 9d ago First seen · 326 lines · 24 tokens per session scan A f385d237376c
transaction-matching is a skill published in the GitHub repository peerjakobsen/smartspender (5 stars, last pushed 7mo ago), licensed MIT. It adds 24 tokens to every session and 3,236 once invoked, about $0.0001 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 2026-08-31.
Other skills, from other repositories
cap-table-analysis
Guides Slice MCP cap-table and ESOP-pool analysis by combining company summary, cap table, ownership, stakeholder, share class, funding round, equity pool, valuation, and securities tools. Use when answering questions about cap table structure, stakeholder holdings, ownership, ESOP pools, share classes, rounds…
workflows-analysis
Guides Slice MCP workflow analysis using workflow list/get-one tools, exercise payment reconciliation, and tax withholding tools. Use when answering questions about grant letters, corporate approvals, share certificate workflows, exercise workflows, pending payments, tax withholding, workflow status, stuck workflows…
board-presentation-analysis
Guides creation of board-ready Slice presentations using MCP tools, especially offered-status grants and ESOP/equity-pool summaries. Use when preparing board decks, board approval packs, grant approval summaries, ESOP pool updates, option pool utilization slides, or board-facing equity recommendations.
equity-compensation-analysis
Guides Slice MCP analysis of equity compensation, grants, options, RSUs, vesting, exercises, forfeitures, PTEP, tax treatments, and valuations. Use when answering questions about employee equity, grant positions, vested or exercisable quantities, exercise price, FMV/409A, option pool consumption, tax treatment, or…
funding-rounds-analysis
Guides Slice MCP analysis of funding rounds, share classes, price per share, primary issuances, secondary transactions, convertibles, SAFEs, option-pool top-ups, total raised, and dilution. Use when answering questions about rounds, Series A/B/C, seed rounds, bridge rounds, investor positions, round history…
waterfall-analysis
Guides exit waterfall analysis using Slice MCP cap-table, share-class, securities, rounds, and ownership tools. Use when answering questions about exit proceeds, liquidation preferences, participating preferred, conversion thresholds, proceeds allocation, MOIC, investor returns, or ordinary/common shareholder payouts.