# Jake Merritt, Independent Developer / Newsletter Operator — read of SearchAgent API, June 18, 2026

> 9 years backend, 2 years solo, running a 4,200-subscriber developer tools newsletter and shipping micro-SaaS on the side.

## How I got here
Someone dropped a link in an indie hackers Discord with the comment "interesting honest take." The word "honest" was doing a lot of work, so I clicked. It was 10:42am. I had just finished my second coffee at the coffee shop I work from after school drop-off.

## What I clicked first
The headline landed: "Stop Your AI Agent from Hallucinating Against Stale Docs." I've built two small agent pipelines in the last six months. Stale docs are a real and specific pain, not a vague one. I was not yet asleep.

Then the subhead: "Purpose-built search API for Claude Code, Cursor, and autonomous workflows." I use Cursor daily. I thought I was about to read about a product I could actually plug in and use.

## Where I paused
Halfway down, the page does something I have never seen before. It pivots from "here's a product" to "here's an idea you can buy to build this product." The tone shifts completely:

"Honest disclosure: we don't have live customers on this idea yet. We shipped the strategy package; you ship the customer conversations."

I had to re-read that twice. So this is not a search API I can use today. It's a blueprint I can buy to build one myself. The page opens like a B2B SaaS homepage and closes like a marketplace listing. Those are two different things with two different readers, and I'm not sure this page serves either one cleanly.

## What I distrusted
Two things kept nagging at me.

First, the scoring section is both refreshing and alarming in the same breath. "-$22,000 Year-1 take-home" and "1 in 8 meaningful-success odds." I appreciate the candor. But I'm evaluating a $99 dossier for a business that Wishdeal's own Fermi math says will cost me money in year 1 and has a 12% shot at mattering. That framing isn't a pitch. It's a disclaimer dressed up as data.

Second: "financial upside: 1/10." That score is on their own axes, using their own model. The team is selling me a strategy package for a business they themselves rated near-zero on financial upside. Pain intensity of 10, financial upside of 1. I don't know how to square that.

The hero copy on the first half of the page is all present-tense active product language: "Returns current, authoritative docs." "Filters SEO spam and outdated references." But there are no customers. That framing needs to be higher on the page, not buried below three sections of product storytelling.

## What would convince me
Show me one person who bought a dossier, built something with it, and hit their first 10 paying customers. Not a testimonial slide. A short written account with a name, the date they bought, what they shipped, and what month they saw the first real revenue. Ugly is fine. Slow is fine. Just real.

Also: what does "1 in 8 meaningful success" actually mean in your model? If meaningful success is $500/month, that's a very different decision than if it means $10K/month. That context is missing and it matters a lot for evaluating whether $99 is a reasonable bet.

## What I'd ask in an email reply
1. The page opens speaking to someone who wants to use a search API and closes speaking to someone who wants to build one. Which reader are you actually writing for? Because right now I think you're confusing both.
2. "You ship the customer conversations" -- what's in the $99 dossier that I couldn't reconstruct from three hours on Google and one HN thread? What's the non-obvious part of the research that earns the price?
3. Has anyone who bought from Wishdeal Studio actually launched a product? If yes, I don't need a referral. I just want a name and a LinkedIn URL so I can do my own digging.

## Verdict: on-the-fence
The pain they describe is real, the unusual honesty about their own scoring is something I actually respect, and the headline earned two minutes of my attention. But the page tries to speak to two buyers at once and I left more confused than I arrived. That is fixable. The more interesting question is whether the math ever gets better than "1 in 8 and you'll lose money this year," and right now the page doesn't give me a reason to believe it does.

---
*Memo by skeptic persona, generated 2026-06-18. Studio breaks own self-grading loop.*