no secret prompt vaults

AI makes money. We show the receipts.

Guides to making money with AI, receipts attached: realistic numbers, honest caveats, and the working prompts sitting right there in the sidebar. Copy, fill in the brackets, run.

stealth-ad-detector.promptChatGPT / Claude
Here is a success story about making money online. Tell me who
profits if I believe it.

Post: [PASTE THE FULL POST]
Top comments, if any: [PASTE]

Run these checks:
1. List every tool, service, or course the post names. For each
   one: does the author benefit if I sign up? Check whether an
   affiliate program exists, whether the author might sell it or
   work there.
2. Check the numbers for internal consistency. Do the views,
   rates, and payouts actually multiply out? Show the arithmetic.
3. Survivorship check: roughly what share of people who try this
   get this outcome? Describe the typical failure case and why
   I'd never see it posted.
4. Rewrite the post's claim as the boring version: "spent X
   hours, made $Y, of which $Z depended on conditions that no
   longer hold."

Verdict: organic experience, stealth ad, or can't tell. One line
on what evidence would change your mind.
Paste any 'here are my honest numbers' success story. Find out who profits from you believing it.

Copied straight from a published guide. Every guide keeps its prompts in an open sidebar like this one.

26
guides published
90
working prompts
0
behind a paywall
0
unsourced revenue claims

Choose your income path

live data

Toolbox

A catalog of AI agent harnesses, rebuilt daily from GitHub snapshots. Grouped by the job each project does — coding agents, frameworks, memory, evals — with the numbers in plain sight.

872 repos tracked · updated daily

Browse the toolbox →

Latest guides

How to Write a Multi-Agent Supervisor: The 39% Rewrite and Why the Savings Weren't Where I Expected

What is a supervisor agent, and when does splitting work into subagents actually pay for itself? I cut my supervisor skill from 3,067 words to 1,872 and found the money was not in reasoning effort. It was in deleting the mandatory root to supervisor to implementer to test-author chain. With the multi agent vs single agent numbers from Anthropic, AWS and a published benchmark that ran the same job both ways.

14 min read

The Claude Code Plugin Marketplace Is a JSON File in a Git Repo. The Only Price Tag in the Spec Is Measured in Tokens.

What is the Claude Code plugin marketplace, how do you add one, and why can nobody sell you a plugin? The full manifest schema has fields for license, category and tags, and no field for money. Checked against the docs, the MCP registry API, and a 66,972-server index on August 2, 2026.

12 min read
2 prompts

'I'm Doing 10x the Work for €40 a Month.' 'Are You Making 10x More Income?' 'Not Yet.'

Two Hacker News threads in the same week: one on a Stanford brief asking what is really happening to jobs, one on companies cutting AI budgets. Between them, 379 comments and a gap nobody closes. Output went up. Income did not. Here's what the numbers people actually cited say, why 'will AI take my job' is the wrong question for anyone selling their own work, and where the money in the threads actually came from.

13 min read
4 prompts

'Vibe Coding Cleanup Specialist' Is an Actual Job Title Now. One Shop Charges $10,000 a Week to Delete Code

A Polish software house posted its price on Hacker News: one week, three senior engineers, $10,000, paid in proportion to how much of the codebase they remove. 240 comments argued about whether anyone would buy it. What the thread shows about who pays to fix AI code, what the work is actually worth, and where the pricing falls apart.

15 min read
4 prompts

How to Run Claude Code From Your Phone: Four Official Paths and the Conditions That Break Every One of Them

Can you run Claude Code from your phone? Yes, and Anthropic now ships four ways to do it. Each one refuses to start under a specific condition: an API key instead of a subscription, Bedrock, a proxy in ANTHROPIC_BASE_URL, zero data retention. That gap is where third-party chat bridges like zarazhangrui/lark-coding-agent-bridge and chenhg5/cc-connect still live. What each option requires, and what you hand over when you put a coding agent in a chat window.

13 min read

Claude Code Swaps the Apostrophe in Your System Prompt Depending on Where Your Traffic Goes

A developer pulled apart Claude Code 2.1.196 and found four different Unicode apostrophes selected by which API endpoint you point at, plus a date format that changes on two timezones. 2,445 points and 677 comments on Hacker News. What the marker actually is, why Claude Code privacy questions keep landing on resellers and proxies, and what it means if you route around the subscription to save money.

13 min read
4 prompts