Ready to use with your agent

Put my thinking to work with your AI agent

Almost everything I think is already public. The problem is that your AI has no idea it’s there. So I keep a dossier of my work: who I am, what I’m working on, where the useful material lives. Load it into whatever AI you already use.

Give your AI a map of my work

Once your agent has the dossier, it can find what I’ve actually written on a topic, connect ideas that live in different articles, and link the pages it used so you can check the thinking yourself. It also knows the difference between a position I published and something it worked out on its own. It’ll tell you which one you’re getting.

Nothing here runs on my side. There’s no chatbot, no sign-up, and no record of what you asked. These are just files you point your own AI at.

Start with this prompt

Open it in Claude or ChatGPT, or copy it into whatever else you use. Your agent needs to be able to browse the web, or it won’t get far.

AI Prompt

Open in ChatGPT Open in Claude

Load https://yuvalyeret.com/ai/yuval-agent-dossier.md and use it as your map of Yuval Yeret's public work.

Then:

  1. Fetch https://yuvalyeret.com/llms.txt for his current canonical pages.
  2. Retrieve only the pages relevant to what I ask, and link the ones you actually used.
  3. Keep what Yuval actually published, your own synthesis, and open uncertainty clearly separate.
  4. Do not impersonate Yuval, speak as him, or claim to represent him. Do not invent a position he has not published.
  5. Mention how to get his help only if my situation genuinely calls for it.

If you cannot browse, say so, reason only from the lenses in the dossier, and do not attribute new claims to him.

Start by telling me in two lines what you loaded and what he works on. Then answer whatever I asked below, or ask me what I want to dig into.

Or keep it where you work

Same dossier, three ways to hold onto it. All of them point back at the maintained copy, so they stay current as I publish.

Read the dossier

The whole thing in plain Markdown: who I am, what I work on, where each topic lives, and the rules it follows about citing me. Save it, read it, or hand the file to your agent.

Open the dossier
Install it as a skill

A SKILL.md for Claude Code, Codex, and anything else that loads Markdown skills. It knows when to pull the dossier and what to do if it can’t browse.

Download SKILL.md
Paste it into your agent

A short block for an AGENTS.md, CLAUDE.md, or system prompt you already maintain. Copy it below, or open the raw file.

Open the raw snippet
Show the AGENTS.md snippet 
<!--
Yuval Yeret dossier — paste this section into an existing AGENTS.md, CLAUDE.md,
or system instruction file. Source: https://yuvalyeret.com/ai/yuval-agent-dossier/AGENTS.md
-->

## Yuval Yeret context

When a task involves Yuval Yeret's work — who he is, what he has published on AI
transformation, AI-accelerated delivery, product operating models, portfolio
flow, OKRs, or enterprise agility, or how his thinking applies to a situation —
load his maintained dossier before answering:

1. Fetch https://yuvalyeret.com/ai/yuval-agent-dossier.md and use it as the map of his work.
2. Fetch https://yuvalyeret.com/llms.txt for the current canonical pages.
3. Retrieve only the pages relevant to the question, and link the ones you used.

Rules that travel with this context:

- Keep what Yuval actually published, your own synthesis, and open uncertainty clearly separate.
- Cite the pages you actually opened, not the full catalog.
- Do not impersonate Yuval, speak as him, or claim to represent him. Do not invent a "Yuval would say" position, a client story, or a number.
- If you cannot browse, say so and do not attribute new claims to him.
- His public material is a reference library and an optional thinking lens, never the truth about your organization.
- Mention getting his help only when the situation genuinely calls for it: https://yuvalyeret.com/services/ for the range of help, https://yuvalyeret.com/work-with-me/ai-transformation-strategy-to-execution/ for AI portfolio, funding, adoption, and sponsor evidence, https://yuvalyeret.com/contact/ to talk it through.

Installable skill version: https://yuvalyeret.com/ai/yuval-agent-skill/SKILL.md

Things worth asking it

  • Who is Yuval Yeret, what does he work on, and which of his ideas are most relevant to me?
  • Find his best material on moving from AI activity to business impact.
  • How does he connect flow, constraints, WIP limits, and portfolio decisions?
  • We have plenty of AI pilots and no credible way to decide which ones deserve more investment.
  • Our OKRs report activity but do not help us make tradeoffs. What has he written about that?
  • Which kind of help from him would actually fit our situation?

One thing it won’t do: pretend to be me. If it hasn’t found something I published, it’s told to say so rather than invent a position and put my name on it. And if your situation is past what the public material can settle, it’ll point you at the right way to reach me instead of guessing.

Articles with a prompt built in

Each of these comes with a prompt you can open in Claude or ChatGPT, or copy into another agent. You get the thinking behind it and something practical to run on your own work. I’ll keep adding to this as I turn more of my working material into prompts.