ingest

ingest is a command for coding agents from CPUtester5465/countbean-plugin. It costs 31 tokens per session (971 once invoked), scanned A, original, MIT.

A command for turning bank statements, card exports, receipts, or plain-text money notes into checked double-entry records in Countbean, a bookkeeping system.

In plain words
What is it for?
Use it to import CSV, OFX, QFX, PDF, receipt images, or described transactions, review proposed entries, and add approved records.
Why use it?
It keeps different source types on the correct import path and flags uncertain details before anything is added to the books.

Command

Part of the countbean plugin — 1 skill, 7 commands, 1 MCP server shipped together

Install

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.

agentmods
npx agentmods add commands/cputester5465/countbean-plugin/ingest
Clone the repo
git clone --depth 1 https://github.com/CPUtester5465/countbean-plugin

Or install countbean, the plugin that ships this one along with the rest of its 1 skill, 7 commands, 1 MCP server.

Wrote 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.

agentmods badge for ingest

README.md
[![agentmods](https://agentmods.dev/badge/commands/cputester5465/countbean-plugin/ingest.svg)](https://agentmods.dev/commands/cputester5465/countbean-plugin/ingest)
Your own site
<a href="https://agentmods.dev/commands/cputester5465/countbean-plugin/ingest"><img src="https://agentmods.dev/badge/commands/cputester5465/countbean-plugin/ingest.svg" alt="Measured on agentmods" height="20"></a>
Per session 31 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 971 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00031 $0.00971
Opus 5 $0.00015 $0.00485
Sonnet 5 $0.00006 $0.00194
Haiku 4.5 $0.00003 $0.00097

Measured 4d ago against content hash f0e7c1899961, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ingest 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 4d 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.

plugin/commands/ingest.md · 68 lines

How it starts

The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Ingest the data referenced by $ARGUMENTS into the user's Countbean book.

$ARGUMENTS may be a file path (CSV, OFX/QFX, PDF, an image of a receipt), a directory of such files, or a plain-language description ("I got paid $4000, spent $200 on software"). If it's empty, ask the user what to import.

CSV, OFX and QFX go through the importer, not through you

propose_transactions parses statements. Do not read a CSV yourself and work out which column is the amount: that answer changes between runs, and the tool's does not.

A RECEIPT IS NOT A STATEMENT and takes the other path. Call stage_receipt with the path first — a receipt is the source document for the entry, and it is kept as evidence outside the book. Read the image yourself (the tool does no OCR, deliberately), then call propose_receipt_transaction with what you read and an honest confidence for each field. It returns a proposal with anything uncertain flagged rather than filled in, and it will tell you BEFORE the write if the account is pinned to another currency or was never opened. Show the proposal to the user, ask about anything flagged, and only then pass its beancount to add_transactions unchanged.

A PDF bank STATEMENT is neither: extract the transactions from it (date, payee, amount, and whether each line is a debit or credit).

  1. Hand the bytes to the tool. Base64 the file and call propose_transactions(content=<base64>, account="Assets:Checking"), naming the account the statement belongs to. It returns proposals, never a write.

  2. Read mapping before you read the numbers. It says which column was taken as the date, the description and the amount; which of the three amount shapes it found (a signed column, separate debit/credit columns, or a running balance); which date format and decimal separator; and whether a positive number was read as money in or money out, with the evidence. If any of it is wrong, call again with the override (amount_shape, columns, date_format, delimiter, decimal_separator, sign, opening_balance) — never by editing the amounts.

  3. Open what it asks for. accounts_to_open and open_directives are ready to pass to open_accounts.

  4. Categorise. This is your job and the tool deliberately does not do it. Every proposal's counter-account is a placeholder (Expenses:Unclassified / Income:Unclassified); replace it with a real account, following the countbean-accounting skill. Keep the import-id: metadata line exactly as given — it is what stops the same statement being booked twice — and keep the ! flags, which mark rows the file itself could not settle.

  5. Preview, then commit. Show the user counts, anything flagged, and a short table. Then call add_transactions with the edited block. The server validates with bean-check and commits to git; if it returns REJECTED, read the error, fix the entries and retry — never leave the user thinking it saved when it didn't.

  6. Report back the commit id, how many landed, how many were skipped as already imported, and anything you flagged. Suggest /countbean:report to see the updated numbers.

Read the full file on GitHub · 68 lines

Changes

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.

  1. 4d ago First seen · 68 lines · 31 tokens per session scan A f0e7c1899961

Subscribe to this mod's changes

ingest is a command published in the GitHub repository CPUtester5465/countbean-plugin (46 stars, last pushed 9d ago), licensed MIT. It adds 31 tokens to every session and 971 once invoked, about $0.0002 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.