Bank reconciliation
Matches the statement to the ledger, including CAMT, BAI2, OFX, and MT940.
Open source, MIT licensed, and run on your own computer. You pay for the setup on your files, not for the software.
The thirty Fairbooks checks are on npm today, version 0.2.0. Each is a small program your AI assistant can call. It reads a folder of exports you choose and writes a report. It has no login to your systems.
npx -y penelope-of-recon mcp --root ./exports
Each check now has a Fair name that says what it does. On npm today it is still published under the name shown on its card, which is what npx uses until the next release. Swap that name for any check below. Needs Node.js 20 or later. Source for all thirty: github.com/fpavia111-creator/fairbooks-agents.
The npm packages are the build of 16 Nov 2026. The fixes from the final pass of 19 Nov 2026, listed on the Fairbooks page, are in the source on GitHub and are not on npm yet. To run the measured version, clone the repository as shown on that page.
Matches the statement to the ledger, including CAMT, BAI2, OFX, and MT940.
Matches orders, receipts, and invoices, including partials.
Reads receipt text and ranks what a person should review.
Masked tax ids and who looks ready for a 1099.
Classifies a card dispute and lists the evidence. It does not refund.
Ties a trial load by company and currency before go-live.
Finds likely duplicate vendors and items for a person to merge.
Plans the count and shows the variance.
Compares budget with actual and packs the evidence.
Checks a percentage against the locked number.
Reviewers, due dates, and segregation of duties.
Reads tool descriptions and lockfiles. It does not deploy.
Tracks expiry from the inventory you export.
Maps the files you have to your own control list.
Compares exported state with the approved baseline.
Reads a contract against your playbook. It does not sign.
Applies your retention rules. It does not delete.
Compares a site capture with your vendor list.
Reads lockfiles and flags a weak license for a person.
The role checklist and the offboarding mirror.
Ages the ball-in-court list from the log you export.
Unused seats against the next renewal you can still cancel.
Estimates a conversion from the scripts you point at.
Checks a changelog against the version you name.
Grades recorded judgments against your rubric.
Reads an exported automation for secrets. It does not run it.
Holds a draft that misses a window. It does not send.
Bad addresses, role mailboxes, and withdrawn consent.
Reads DMARC and a saved header.
Contrast, focus, and labels from the static page.
The checks in every other product are installed as part of a setup and are not on npm yet; their source is public on GitHub. They are named the same way, the product and then the job, such as fairlease-security-deposits. Ask for the list for your file.
Each product's checks, test files and demo are public on GitHub under the MIT licence. The 14 measured products also include their accuracy scripts. Clone one, run npm install, then its demo: it runs on made-up files with no network. The exact commands are on each product page.
Mythos Machines also publishes open-source tools for people who build with AI coding assistants. The cores are free and some have paid add-ons. They are a separate line from the Fair Agents checks and live on our sister site: g6daddy.com/tools.html.
Turns a NetSuite saved search, an Okta user list, a Stripe dispute export, or a bank OFX file into the layout the checks read.
Lets a check keep notes about its own job. Anything it proposes to learn waits for a person to accept it.