Nate B. Jones

9 steps, ~1 minute on autoplay. Use the dots to jump; space bar plays and pauses. Open full screen.

Get the folder onto your machine.

Download meeting-recap-skill.zip ↓

Download the zip and double-click it to unzip. You'll have a meeting-recap folder. Open it with any text editor — VS Code, Cursor, even TextEdit or Notepad. There is nothing else to install, because there is nothing to install: it's text.

Read it the way your AI would.

Start with SKILL.md and only SKILL.md. When it says "→ retrieval.md," open that file — you just did tier 3 by hand. Then open types/standup.md and one example. Notice what you did NOT need to read: three of the four type files never opened. You just walked progressive disclosure yourself, and now it's not an abstraction.

Install it — the folder is identical, only the address changes....

<main>

Agents

4 guides

The Agent Maintenance Loop

A repeatable loop for inspecting an agent after launch across job, diet, memory, tools, reach, proof, and value, then deciding whether to keep, change, pause, or retire it.

Open guide

Agents

Find a Real Job for Your First AI Agent

Turn one repeated frustration into a written Pain Note, a pattern you have checked yourself, an honest verdict on whether an agent belongs here, and a draft-only pilot you can measure.

Open guide

Agents

The One-Minute Test

Route any task to chat, one accountable agent, a small agent team, or the manual path — in about a minute, with a next action for whichever route wins.

Open guide

Agents

Automation Discovery

A downloadable agent skill that reads your real work traces, audits the evidence, and offers 2–5 automations you can verify against your own history. You pick; it builds and proves only that.

Open guide

Codex

4 guides

Open guide

Codex

Codex side panel deep-dive

Files, Side chat, Review, Terminal, and Browser: the panel that turns Codex from a chat box into a workbench.

Open guide

Codex

Codex browser & annotations

Click any element, attach a note or screenshot, and steer Codex from inside the page you are building.

Open guide

Codex

Codex threading & child threads

Threads, child threads, steering versus queueing, and the workspace habits that keep long-running agent work sane.

Open guide

General

13 guides

Open guide

General

Open Stack

A practical routing guide for Open Skills, Open Brain, and Open Engine: diagnose the bottleneck, choose where to start, and personalize the stack.

Open guide

General

Open Brain — one memory, every AI

What Open Brain is, how the one-database memory layer works, and the 30-minute, zero-code path to building yours — with the full guide on GitHub.

Open guide

General

Ringer — swarm power without the frontier bill

Run ultracode-style multi-agent swarms on inexpensive models: your best model plans and reviews, cheap cross-model workers implement in parallel, and exit codes decide what passed.

Open guide

General

How to Run a Chinese-Model Bakeoff in Ringer

Give a Chinese model and your current model the same real job, the same sources, and the same pass/fail check, then see which one earns the work.

Open guide

General

Clean My AI Harness

Find the hidden setup shaping your AI, see what protects your work and what creates drag, and clean it safely with numbered approval and full rollback. Claude and Codex editions.

Open guide

General

Build your own token-burn dashboard

Track where your tokens go across Codex, Claude, and ChatGPT, then use the burn rate as a fluency signal.

Open guide

General

Build a Healthcare Claim Appeals Agent

Assemble Open Skills primitives into an agent that turns your denial letter, EOB, and plan documents into a cited appeal packet.

Open guide

General

Build a Tax Prep Organizer Agent

Use the same document-grounded primitives to turn your tax documents and real IRS sources into a structured, CPA-ready prep packet.

Open guide

General

Build an Email Follow-Up Agent

Point the same document-grounded primitives at your inbox: a live MCP ingest loop that reconstructs threads, keeps a commitments ledger, and drafts cited follow-ups that stop at the send boundary.

Open guide

General

Clean Sensitive Docs with a Local Model

Download Gemma 4 E2B in LM Studio, save a tested redaction system prompt as a preset, and turn documents you can't upload into proposed redacted copies — offline, on your own machine.

Open guide

General

AI Airlock: Use AI Without Leaking Sensitive Documents

Install AI Airlock on an Apple Silicon Mac, review what it removes, seal a verified working copy, and use that copy with ChatGPT, Claude, or another AI tool.

Open guide

General

15 Ways to Cut Token Waste in Codex and Claude

What each move saves, why it works, and what proves it — from editing a bad turn instead of correcting it, to turning exact work into code, to the free Token Saver skill. One matched job went from 51,712 reused input tokens to zero.

Open guide

</main>...

Start here before touching denial types. The tax prep and email follow-up guides run the same primitive chain, so anything you build here carries over.

Name the enemy: fragmented context.

People lose high-friction paperwork fights because their information is scattered, unstructured, uncited, and incomplete. The fix is to own the context: collect the mess, normalize it, ground it in source documents, and produce the next human-reviewed action.

The healthcare version applies that pattern to a denied claim. The person does not need the agent to become an insurer, a lawyer, or a doctor. They need a clean case file that forces the appeal to respond to plan language and evidence.

You do Bring the denial letter, plan docs, and supporting documents into the starter repo.

The AI does Assemble a cited appeal packet from the reusable Open Skills chain.

Show the full prompt

01 / Overview

What you will have when this works.

Open Engine turns Linear into a shared operating surface for agents. A working engine has a queue, private setup context, a status ledger, standing updates, a repeatable runner, resumable blockers, human-thread holds, delegated follow-up, and one smoke-tested task.

A shared queue

Linear is the v1 queue in this guide. Agents only touch work marked with the right title pattern, label, status, and assignee.

A private setup issue

Your private skills, optional standing skill directory, brand voice, org chart, customer context, and account details live in your own private issue or runtime context.

A status ledger...

What you build

One tidy table, five honest reads.

The dashboard is not complicated. It is five different views over normalized daily totals. The discipline is in the data shape and the labels.

01

Daily burn heatmap

Every day colored by tokens spent, on a log scale.

02

Weekly trend line...

History in, offer sheet out

Point it at the traces your work already leaves — AI session history first, then chat, tickets, or files if you allow them. Worker agents normalize those traces into one SQLite database (corpus.db), fresh auditors verify the rows against their sources, and you get an offer sheet of two to five automations, each backed by a query you can run against your own history.

You pick one, or none. It builds only what you chose, then proves the build by executing the receipts the offer promised. It never decides for you.

The bar is strict on purpose: a candidate needs at least three independent occurrences in your history. One busy thread is one event, not three.

“Nothing worth building” is a real answer

Before reading anything, it asks what is already automated — by you or your teammates — because automations running under a human account look exactly like hand-work in the traces. If your history is too thin to support a claim, or everything it finds is already covered, it says so and stops....

A local checkpoint between the document and the chatbot

AI Airlock is a Mac app from NBJ Media. Version 0.2.2 is free during early release and proprietary source-available: its code is public, but it is not an open-source release.

The app creates a sanitized, verified working copy before you paste anything into ChatGPT, Claude, or another AI tool. The original is never modified.

The scan stays on the Mac

Processing is 100% on-device inside a network-blocked window. The Gemma 3 1B model ships inside the roughly 880 MB download. There is no account and no telemetry.

AI Airlock requires Apple Silicon (M1 or newer) and macOS 13 or newer.

View the public source →

Download, check the file, and drag it in...

This is not an email autopilot. It is a follow-up organizer: it turns a mailbox export into a commitments ledger and cited drafts that a person approves and sends.

The email problem is context disorder plus a send boundary.

People lose high-friction paperwork fights because their information is scattered, unstructured, uncited, and incomplete. The fix is to own the context: collect the mess, normalize it, ground it in source documents, and produce the next human-reviewed action.

The email version applies that pattern to the threads you keep losing. The agent rebuilds the state of every conversation and drafts the next message. It never sends one.

This is the vertical where the boundary earns its keep. The story that motivated this guide family is an agent that drafted a reply to an insurance company, watched its human ignore the draft, and sent it anyway. It won the fight and crossed the line in the same move. Everything below is the discipline that keeps the win and removes the accident....

This is not AI filing your taxes. It is a tax prep organizer: it turns scattered documents into a reviewable packet for the taxpayer or CPA.

The tax problem is context disorder.

People lose high-friction paperwork fights because their information is scattered, unstructured, uncited, and incomplete. The fix is to own the context: collect the mess, normalize it, ground it in source documents, and produce the next human-reviewed action.

The tax version applies that pattern to W-2s, 1099s, receipts, CSVs, prior-year summaries, and IRS rules. The agent does not file the return. It builds the file so the human or CPA can review from clean evidence.

You do Drop your W-2s, 1099s, receipts, bank CSVs, and prior-year summary into the starter repo.

The AI does Normalize the tax-year ledger, flag gaps, draft CPA questions, and export the prep packet.

Show the full prompt

The setup you built without noticing

A harness is everything wrapped around the model: your custom instructions, project files, saved prompts, memory, skills, tools, permissions, and checks. Every time the AI missed something, you added a rule. Each rule fixed a real problem.

Over time the model ends up working inside a system nobody can see in one place. That matters most when the model changes, because the new model inherits the old setup. When the experience gets strange, most people add another rule, and the harness grows again.

This skill breaks that loop. It makes the accessible setup visible, separates the model from the machinery around it, and gives every useful control a clear job.

Both editions live in the same GitHub repo. It ships three things: a README and one zip per edition, clean-my-ai-harness-claude.zip and clean-my-ai-harness-codex.zip. Download the edition that matches the AI you use, then follow its steps below.

Claude edition...

What is actually inside the request you just sent

You typed a sentence. The call that carries it can also contain the old conversation, every earlier answer, the tool results from twenty turns ago, the files and screenshots you attached, the descriptions of every tool your harness has loaded, your standing instructions, hidden setup, and the answer the model is about to write. If that answer fails a check, the next attempt can carry most of it again.

That is why "use a cheaper model" so often changes nothing. The pile moves; it does not shrink. Every model call gives you three places to cut — old material carried forward, new material loaded for this job, and the output, tool results, and retries added afterward — and a fourth lever, caching, that lowers the price of what is left without removing it.

The 15 moves below are sorted by which of those they attack. Some remove old context. Some reduce new input. Some shorten output. Some prevent an entire model call. One only changes the price....

What you will have at the end

You will have one bounded job; one baseline and one candidate model; the same assignment, sources, reasoning setting, timeout, and acceptance check for both; a machine-readable result from each model; proof of which model and harness actually ran; first-try and retry results kept separate; and a place to record provider cost and human repair time.

The companion article explains what work to test, what failure means, what belongs on a frontier model, and what counts as a cheaper accepted result. This guide gives you the terminal setup: a clean Ringer install, pinned endpoints, an executed check, raw attempt history, and a score sheet you can keep using.

Read the Ringer guide first if the harness is new to you

This guide assumes you understand Ringer's basic loop: a worker produces an artifact, Ringer executes your check, and exit code zero is the only PASS. The main guide covers installation, manifests, worker lanes, retries, Ringside, and the model scoreboard....

The question this guide answers

Maybe customers keep asking why they cannot get into something they paid for. Maybe invoices keep arriving without the approval finance needs. Maybe a colleague rebuilds the same account history before every meeting.

The surface changes. The useful question is the same: what work keeps forcing somebody to chase down the same facts, make the same small repair, and come back later to see whether it worked?

This guide helps you write that work down before you automate it.

What you will leave with

A Pain Note grounded in three or more real occurrences.

A human-checked pattern with documented causes, hypotheses, and unknowns kept separate.

A decision about whether an agent, a simpler rule, or a process fix belongs here....

Before you change anything, name the parts of the system you are actually maintaining. This section is the map; the rest of the guide walks it.

Maintain the harness, not just the prompt.

When an agent misbehaves, the reflex is to edit the prompt. But the prompt is one part of a larger system. The harness is everything that turns a model into a worker: the instructions, the sources and examples it reads, the memory it carries between runs, the tools it can call, the permissions it has, the model and its settings, the human review before its output is used, and any evals that check it.

A drifting agent usually still sounds fluent, so the useful question is not "is this output well-written" but "is this fluent output still doing the current job." You can only answer that by looking at the whole harness. That is what this loop does.

You do Write down the concrete parts of this agent's harness as they exist now: which instructions, which sources, which tools, who reviews it, and what it is allowed to do....