Compounding Offer Audit: score an offer on six tests, find its weakest point, change one thing
by PaperChaseWebb
Score each thing you sell from 0 to 3 on six tests: recurring, productized, systemized, owned demand, compounding, and whether growing it would make the owner's week better. Every score needs a checkable fact and its source, or the word unknown and the fact you are missing; no 3 on the owner's say-so, and no total. The audit names the weakest test, treats a low life score as a veto on growth, and proposes one change with a success test and a date. Every change waits for the owner's yes. Fifteen read-only checks run on the folder.
- Price
- $19
- Difficulty
- Intermediate
- Setup
- 1 hr
- Format
- .md file
compounding-offer-audit-score-an-offer-on-six-tests.md · 110 KB · instant download after checkout
Want this built for you instead? We can build it for you.
What you end up with
For every offer you sell, you know its weakest test, the fact behind each score, and the one change to make next; and no change starts without the owner's yes.
What you get
- 21 sections in one .md file, with a one-line unpack
- The six tests with a written 0 to 3 scale for each, and the rules: a fact and source per score, unknown instead of a guess, no 3 on say-so, no totals, life as a veto
- What to do about each kind of low score, as one change with a success test and a date, and a one-page worksheet
- The audit folder: audit.json, offers, scores, actions, approvals and done, with blank templates
- offer.py: gate (exit 0, 11, 12), a one-page audit.md, owner approval tied to a sha256, verify and one use per yes
- offer_check.py: fifteen read-only checks, with a seeded test audit (15 findings, then 0) and the unit tests
- A skill file for Claude Code, Codex and Hermes, an AGENTS.md snippet and a Grok Bot rules block
- Ten prompts, a failure table, a worked example on a made-up studio, and a look at other offer audits with sources
Who it is for
Owners who sell two or more things and want to know which one to fix first, and what to change, from facts rather than a feel-good total.
Works with
Solid chips: the agent is named in this listing's own tools list. Dashed chip: A skill file for Claude Code, Codex and Hermes, an AGENTS.md snippet and a Grok Bot rules block; the scripts were tested on Linux with Python 3.13, the agent files were not loaded inside each harness and the audit was not run on a real business.
Tools needed
Cost to run: The pack is the only purchase. The scripts use no model tokens; scoring with an agent uses its usual tokens.
Time to result: A first audit of one offer in an afternoon, with invoices and calendar open
How it works
Run the tests
Unpack the pack with the one-line command in section 8 and run sh test_all.sh.
Expected: 15 findings on the seeded audit and 0 on the fixed one, 15 of 15 single faults caught alone, checker tests 50 passed, offer.py tests 37 passed, exit 2 on a missing folder.
Fill the worksheet
Pick one offer and fill the worksheet in section 7 with your invoices and calendar open. Do not add up the column.
Expected: Six scores, each with a fact and a source or a named missing fact, and the weakest test circled.
Set up the folder
Run offer.py init, fill audit.json and offers.csv, then use prompts 2 to 4 to score each active offer and propose one change on its weakest test.
Expected: The checker shows 0 findings apart from O15 before ratification.
Get the yes
The owner ratifies the rules and approves each change at their own terminal. Run offer.py report and read audit.md together.
Expected: A one-page audit with six scores per offer, no total, the weakest test named, and every change tied to the owner's yes.
Sample prompts, free to use
Score offer o1 as audit A1 on all six tests using only my invoices folder, my calendar export and our written procedures. One sentence of evidence per test, with its source. Write unknown where you do not have the fact, and name the fact. No 3 on anything I only told you.
Prompt 2 of 10.
For each active offer, tell me the weakest test and propose one change for it in actions.csv, with a success test I can check and a due date inside the window. If life is low, the change is about life.
Prompt 4 of 10.
Draft a follow-on plan for offer o1's buyers, offered at handoff: what it covers each month, what it does not, and how a client cancels. No lock-in, no annual auto-renewal, no exit fee. Draft only; I will decide whether to offer it.
Prompt 5 of 10.
Free preview of the file
# Compounding Offer Audit: score an offer on six tests, find its weakest point, change one thing Version 1.0, 2026-10-29. By PaperChaseWebb. A scoring audit for the things you sell. Each offer gets six scores from 0 to 3: how recurring it is, how productized, how systemized, how much of its demand you own, whether each sale makes the next one easier, and whether growing it would make the owner's week better or worse. Every score carries a fact and where the fact came from, or the word unknown and the fact you are missing. There is no total. The audit names the weakest test and the one change to make next, and the owner says yes to that change before anything moves. Plain files, a script that gates what an agent may do and writes a one-page audit, a read-only checker with fifteen checks, a worksheet, templates, a seeded test set, ten prompts and the tests, all Python 3 standard library. Works with Claude Code, Codex and Hermes through a skill file and an AGENTS.md snippet, and with Grok Bot through a rules block. Tested as described in section 13: everything ran on Linux with Python 3.13, including owner approvals typed at a pseudo-terminal. The skill file, snippet and rules block were not loaded inside the harnesses for this release, and the audit has not been run on a real business. Each is marked where it matters. Most offer audits you can download end in a single number. Add up ten categories, get 71 out of 100, feel fine. The trouble is that a strong price page can hide
Test record
- Date
- Oct 29, 2026
- What was run
- offer_check.py on a seeded audit and a fixed copy, each planted fault alone, a missing input, and the unit tests. Linux, Python 3.13.
- Result
- 15 findings on the seeded audit, 0 on the fixed one; 15 of 15 planted faults caught alone; exit 2 on missing input; nothing written. 50 checker and 37 offer.py tests passed, 0 failed.
- Not run
- A real business; the fixture businesses are made up. The agent files inside Claude Code, Codex, Hermes or Grok Bot. Windows and macOS.
Copied from this item's dated test notes. What was not run stays in.
Worth knowing
- The audit has not been run on a real business; Fennick Studio, Marlow Tidy Books, their people and every fact about them are made up.
- The pack makes no claim that a higher score means more revenue, or that any change will earn a particular amount. It does not set prices or forecast income.
- The skill file, AGENTS.md snippet and Grok Bot rules block were not loaded inside Claude Code, Codex, Hermes or Grok Bot. Linux only; Windows and macOS were not tested. The scripts open no network connection.
- The approval step refuses piped input and the environment variables Claude Code, Codex, Hermes and Grok Bot set. That stops an agent approving its own change; it is not a lock against someone with the same account.
- On 2026-10-29, on Linux with Python 3.13: offer_check.py gave 15 findings on the seeded audit (each check once) and 0 on the fixed one, caught 15 of 15 single faults with only its own check, wrote nothing, and exited 2 on a missing folder, missing scores.csv or broken audit.json; checker tests 50 passed, 0 failed; offer.py tests at a pseudo-terminal 37 passed, 0 failed.
Questions
Why no total score?
A total lets a strong price page hide an offer only the owner can deliver. The audit keeps six scores side by side and names the weakest one, which is where the next change goes.
What if I do not know a score?
Write unknown and name the missing fact. Finding it becomes the first change. A guessed score is worse than an honest unknown.
How is this different from Workflow, Not Wrapper?
Workflow, Not Wrapper helps you design what you sell, with the offer ladder and a one-table model check. This pack scores what you already sell, with evidence, and picks the next change. They link to each other rather than repeat.
Will an agent change my prices or message my clients?
No. Contacting clients, changing live prices, posting, spending and signing are gated never: the agent drafts, the owner does those by hand.
Details
Version 1.0 · Updated Oct 29, 2026 · Markdown version of this page
License: Use for your own business and for audits you run for clients. Do not resell or republish the pack itself. Download terms
Related
Compounding Offer Audit: five rules and a five-check script (free sample)
From the Compounding Offer Audit: five rules for scoring an offer on six tests (one owner, six scores and no total, a fact for every score, no 3 on say-so, one change on the weakest test), and a read-only script that checks one small file for them.
Workflow, Not Wrapper: test an AI product idea, shape the offer and set the price
A prompt pack and method for anyone building or selling on top of a model. A seven-score absorption test, a one-page workflow product spec, a demo gap check, a problem mining loop, a five-talk buyer script, a channel translation table, a model column check, a true cost rebuild at three usage levels, an offer ladder with pricing rules, a price anchor audit, a check for copied success stories, a fidelity gate for generated output, a failure table, a worked example, an idea folder with a tested checker, a rules block and 19 prompts. What changed in v1.1: the demo gap check, buyer talks, true cost, price anchors and the checker are new.
Problem Mining Loop: real complaints, graded and counted, before you build anything
Find out what one kind of buyer complains about, in their own words. Every complaint is a row with the link, the date posted, the exact quote and a salted key in place of the poster's handle, graded gripe, pain, budget or buyer. Clusters count independent posters, not posts, and leave out prompted answers and your own accounts. A cluster is proposed only when enough people at pain or higher, from more than one source, say so and one shows money. Public pages only, robots.txt and terms checked, GET only, and nobody in the community is ever contacted. Proposals wait for your yes at your own terminal; fifteen read-only checks run on the set.
Source Shelf: a short, owner-approved list of industry reports your team reads, and a way to read them
Keep a small set of major industry reports as your team's default evidence base: each with its edition, publish date, link on the publisher's own site, the date someone last looked for a newer edition, and a note on what the publisher sells. Read any one of them against three questions (what is changing, who is struggling, what could help them), with word-for-word quotes and page numbers, the third answer marked as your inference, and every figure matched to its quote. Every add, drop and replace waits for the owner's yes. Fifteen read-only checks run on the folder.
Evidence Standard: what counts as proof for each kind of agent work, who checks it, and what is refused
Agents report done from the plan as often as from the result. Five plain files write down, for each class of work, the proof that counts, who checks it (never the producer), what is refused and how long proof is kept. Agents log each claim with its evidence, seal it by sha256 and report in the words evidence.py say allows: done, partly done with the gap named, waiting on the owner, or reported, not verified. The agent's own summary is never proof. Fifteen read-only checks run on the standard and the claims ledger.
Agent Readiness Self-Check: score one workflow before you hand it to an AI agent
A free self-check for agencies and small businesses. Name the decision, then score one workflow across eight areas (workflow mapped, data access, approval gates, owner per agent, handoffs, cost tracking, security and vetting, measurement) with an evidence rule, three gate rules and a skeptical pass, and read your band and the next steps it allows. v1.1 starts with the decision the score will change, scores unknowns as unknown, adds rules for conflicting facts and lookalike public records, ties every band to the next steps it allows with the Production Readiness Audit before go-live, makes owner approval concrete (prepared and approved status, a one-reply approval line, decisions the agent must leave open, a done test), adds agent files for Grok Bot, Claude Code, Codex and Hermes, a second worked example (missed call to booked job), a readiness folder with a fifteen-check read-only checker, a front door test with four questions to ask anyone selling you an agent install, and a plain note on how it was tested. Still free.
Startup Credits Sweep: an eligibility matrix, honest applications and an expiry ledger
Sweep startup programs for cloud, AI API and SaaS credits from your own facts. Each fact names its record; each program is read on its own page with the date; each requirement is marked meets, fails or unknown against one fact, and the matrix sets the verdict. Application drafts cite every fact and say it word for word, so nothing is stretched or invented to fit a form. A ledger counts credits only after the approval notice and reminds you before each grant expires. Every application waits for the owner's yes, and the owner submits it by hand. Fifteen read-only checks run on the folder.