Live 175 operators 233 cited frameworks

“Act like an expert” is a costume.
Gavel hands you the playbook.

Ask a coding agent to price your product and it builds the expert out of training data. You get the internet's average of that person, frozen at the cutoff. Gavel hands it the real framework instead. It names the operator and links to where they said it.

Give this to your agent

It fetches the skill and saves it. From then on it calls Gavel on its own.

$ set up https://usegavel.com/SKILL.md

Three reasons the impression fails

Your agent already does this. It picks a name, rebuilds the person from weights and ships the call.

01

You get the average, not the person

Your agent builds April Dunford out of everything it read. Most of that is mediocre, and all of it stopped at the cutoff.

Most of the product management contenting existing on the Web is of terrible quality... LLMs were trained on all of that data, and, therefore, produce very poor product decisions.
Comment on Lenny’s Podcast · Product Management Is Dead · 38 likes

Gavel retrieves what the operator said. It quotes the line and links the source.

02

The hard part is knowing who to call

Say churn is climbing. Jason Lemkin calls it a sales problem, so you sold to the wrong customer. Bob Moesta calls it a jobs problem, so the customer's job changed. Those point at opposite fixes. Pick wrong and you burn a quarter.

Your agent averages the two into a paragraph that offends no one. Gavel routes to the operator who solved it, then shows you the fork.

03

It does not know your business

A persona built from weights advises a stranger. It has no idea that you are pre-revenue and that outbound flopped.

Your agent holds that context already. Pass it in the query and Gavel routes on it.

One call answers all three

The agent sends a decision. Gavel returns the operators, the frameworks and the place the two of them split.

$ curl -s "usegavel.com/api/v1/route_expert.txt?q=first+10+customers,+no+audience"

[1] Paul Graham
FRAMEWORK Do Things That Don't Scale
WHY       Startups don't take off by themselves. Recruit the
          first users by hand.
QUOTE     "The Airbnb guys would go out and do that, and that
          was clearly something that did not scale."
SOURCE    youtu.be/IjPDIjge81o?t=206 (3:26)
PAGE      usegavel.com/paul-graham/do-things-that-dont-scale

[2] Alex Hormozi
FRAMEWORK Core Four Customer Acquisition
SOURCE    youtu.be/CaiLcj6tzBQ?t=84 (1:24)

DISAGREEMENT
          Hormozi picks one channel and runs it hard. Graham
          says stay manual and unscalable until it hurts.

Your agent has opinions. It doesn't have receipts.

We put the same ten founder decisions to two frontier models and to Gavel.

Then we opened every URL that came back.

0
Gavel sources that failed to resolve, out of 35.
8
URLs the two models invented. The authors are real. The pages are not.
31
Gavel sources that open on the exact second. The models returned none.

A model can name an operator. It cannot show you where that operator said it. That gap is the whole product.

Who your agent will be citing

233 cited frameworks are callable today, from 175 named operators. 162 of them open on a timestamp and the rest land on the book or the talk the claim came from. The catalogue holds 280 entries; the other 47 cite a source with no link, so the API holds them back. When nothing fits, the call says so instead of guessing.

Three ways in. Same corpus, same citations.

Skill file

One line into your agent's chat. It installs itself and knows when to call.

$ set up https://usegavel.com/SKILL.md
Agent rules

Paste one block into AGENTS.md or CLAUDE.md. Every run in that repo becomes a caller.

## Product and business decisions

Before proposing pricing, positioning,
or go-to-market changes, run:

  curl -s "https://usegavel.com/api
  /v1/route_expert.txt?q=<decision>"
Raw HTTP

Three GETs and no key. It works from a shell, a cron job or your own tool definition.

GET /api/v1/route_expert
GET /api/v1/framework/:id
GET /api/v1/catalog

The three calls

GET /api/v1/route_expert

Give it a decision in plain language. It reads the whole catalogue and returns up to three ranked operators. Each one carries the framework it applies and why that fits. It also names the tension between them.

?q=should+I+raise+prices
&format=txt   # ledger, human + agent
              # omit for JSON
GET /api/v1/framework/:id

Already know the page you want. It returns the quote, the timestamped link and the disagreement. No model sits in the path, so it cannot stall or invent.

/api/v1/framework/
  april-dunford/
  competitive-alternatives.txt

GET /api/v1/catalog returns every page Gavel can cite, as JSON. It never emits colour, so piping into an agent is lossless. Add &full=1 for untruncated quotes.

Questions your agent will ask first

Point your agent at it and see.

8 free calls a day, no account and no key.

$ set up https://usegavel.com/SKILL.md