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 lilfroog/qbbridge --skill qbbridge-bookkeepinggit clone --depth 1 https://github.com/lilfroog/qbbridgeWrote 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/lilfroog/qbbridge/qbbridge-bookkeeping)<a href="https://agentmods.dev/skills/lilfroog/qbbridge/qbbridge-bookkeeping"><img src="https://agentmods.dev/badge/skills/lilfroog/qbbridge/qbbridge-bookkeeping/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/lilfroog/qbbridge/qbbridge-bookkeeping"><img src="https://agentmods.dev/badge/skills/lilfroog/qbbridge/qbbridge-bookkeeping.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 | $0.00089 | $0.02226 |
| Opus 5 | $0.00044 | $0.01113 |
| Sonnet 5 | $0.00018 | $0.00445 |
| Haiku 4.5 | $0.00009 | $0.00223 |
Grade A, and why
qbbridge-bookkeeping 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 10d 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 — 183 lines — stays where its author put it; the contents beside it link to each section on GitHub.
QBBridge bookkeeping
This skill is the judgment layer on top of the QBBridge MCP server. The
server (qb_* tools) only exposes safe, generic QuickBooks Desktop
operations -- it has no opinion about which vendor maps to which account,
when to trust a match, or how confident is confident enough. That
judgment lives here. Follow it exactly; it encodes rules that were learned
the hard way, by getting them wrong first against real client data.
The one rule everything else follows from: never fabricate a vendor, account, or classification. Every decision must trace back to either existing evidence in the company file, or an explicit instruction from the user in this conversation. If neither exists, flag it -- don't guess.
0. Before doing anything
- Call
qb_company_info. Confirm out loud, in your response, which company file is open and that it matches who the user says they're working on. If the user hasn't said which client/file this statement is for, ask -- do not infer it from the file being open. - Save that file path. Pass it as
expected_company_fileon every write tool call for the rest of this task. This is the single check that prevents posting to the wrong client's file, and it costs nothing to include every time. - If the user gives you a statement but QuickBooks isn't open, or is open on a file they didn't name, stop and ask -- don't guess which file they meant.
1. Parse the statement into normalized transactions
Whatever the source format (CSV, QBO/OFX, a PDF you're transcribing, a
screenshot), produce a plain list of (date, description, amount) before
doing anything else. Two things to get right here:
- Sign convention. Bank/card statements are not consistent about whether a purchase is positive or negative. Look at 2-3 obviously-a- purchase lines and 1-2 obviously-a-payment lines to determine the convention for this statement before processing the rest. State the convention you inferred back to the user in your summary so it can be caught if wrong.
- Don't collapse near-duplicate lines. Multiple charges from the same vendor on the same day for the same amount are usually real, separate transactions (e.g. several equipment purchases in one visit), not a parsing artifact. Only treat lines as duplicates per the dedup rules in section 3, never by eyeballing "these look similar."
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.
- 10d ago First seen · 183 lines · 89 tokens per session scan A e134d7468851
qbbridge-bookkeeping is a skill published in the GitHub repository lilfroog/qbbridge (1 stars, last pushed 20d ago), licensed MIT. It adds 89 tokens to every session and 2,226 once invoked, about $0.0004 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
sector-rotation
An analysis framework for comparing industries in the Chinese A-share stock market, using business conditions, price momentum, valuation, and money flows. It produces rankings and higher- or lower-allocation suggestions.
strategy-pivot-designer
Detect backtest iteration stagnation and generate structurally different strategy pivot proposals when parameter tuning reaches a local optimum.
twitter-reader
Read Twitter/X for financial research using opencli (read-only). Use this skill whenever the user wants to read their Twitter feed, search for financial tweets, view bookmarks, look up user profiles, or gather market sentiment from Twitter/X. Triggers include: "check my feed", "search Twitter for", "show my…
chenhao-limit-up
A framework for judging Chinese A-share stocks that have reached the daily price-rise limit, using market mood, sector leadership, and trading momentum.
trading-risk-gate
Unified pre-trade safety gate: Ruin check (Law #1), ergodicity audit, and win-rate dominance validation. Absorbs: ergodicity-check, law-of-ruin, win-rate-dominance.
furusato
A Japanese hometown-tax donation manager for furusato nozei, a system where donations to municipalities can qualify for an income-tax or local-tax deduction. It reads donation receipts, stores donation records, and calculates deduction limits.