A practical workflow for the next decision
A famous framework becomes useful when it changes what you do on Monday. This resource pairs a link to Sam Altman’s actual Startup Playbook with original worksheets and Claude skills for testing your startup’s assumptions. You bring the company context; the pack helps turn it into a decision brief and a short list of experiments.
The original playbook was published through Y Combinator in 2015. It covers the idea, team, product and execution, with sections on topics including hiring and fundraising. Our download is an independent implementation toolkit. It does not reproduce the book, apply an official YC score, predict acceptance or claim to reveal private investment criteria.
Use it when you have a project, early customers or a pitch and want to find the next important uncertainty. It is also useful for a strategy team evaluating a new venture. You do not need to manufacture a complete business plan before starting: five honest facts and three unresolved questions are better inputs than thirty pages of confident prose.
Free download. Choose an individual skill below if you prefer to start with one workflow.
How to work through the pack
Write the decision you are trying to make: build an initial product, choose a customer segment, prioritise a product change or prepare for fundraising. The same company may need a different review for each decision.
Collect evidence you already have. Customer notes, usage records, paid pilots and product demos carry different weight. The assessment skill separates observations from founder beliefs and flags where several claims rely on the same weak source.
Choose the uncertainty that could change the decision. Design a small test with a named audience, observable result, time limit and threshold set before the test begins. A marketing activity is not an experiment until a result could make you change course.
Close the week with a decision log. Record what happened, what surprised you, which assumption changed and what you will do next. Keep unresolved questions visible rather than replacing them with a polished narrative.
Bring the right inputs
- A one-paragraph company description and the decision you want to make
- Customer notes, product or usage evidence, with dates and permission to use it
- Your current constraints: time, budget, team capacity and runway
Try this first
Use the startup-evidence-review skill on this company brief and customer evidence. Separate facts, assumptions and unknowns. Select the three most consequential gaps, then use experiment-design to propose one test I can finish next week. Do not invent traction or score my chance of getting into YC.Install a skill and run one task
- Download the pack, or choose one individual skill ZIP from the directory below.
- Upload that skill ZIP using the skill settings available in your Claude account. Keep the skill folder and its supporting files together. For Claude Code, place the skill folder in your project’s .claude/skills/ directory.
- Share the context, evidence and output you need. Ask Claude to use the selected skill, then review its answer before acting on it.
You can also attach selected Markdown instructions to a Claude Project as reference material. Account features and upload limits can differ; the archive includes a README and the complete skill folders so the instructions remain usable.
Free download. Choose an individual skill below if you prefer to start with one workflow.
The 3 skills in this pack
01
Experiment design
A startup assumption needs a concrete customer or product test. Produces A test brief with audience, budget, evidence threshold and stop rule.
Output: A test brief with audience, budget, evidence threshold and stop rule
Download this skill (.zip)02
Founder weekly review
A founder wants to convert the week\u2019s evidence into a focused next week. Produces A decision log, one priority and an owned seven-day plan.
Output: A decision log, one priority and an owned seven-day plan
Download this skill (.zip)03
Startup evidence review
A new venture needs an honest assessment before the next decision. Produces A sourced decision brief with the three most consequential evidence gaps.
Output: A sourced decision brief with the three most consequential evidence gaps
Download this skill (.zip)Review the work, then tell the story
The worksheets and skills are Oria’s original work, not official YC material or an endorsement. Read the original playbook alongside them.
A review can organise evidence and reveal assumptions. It cannot establish customer demand, founder quality or investment merit without real-world evidence.
Once the evidence and conclusions are reviewed, turn them into a page brief: one main point, the supporting facts, the visual you need and the source notes. Oria helps turn that brief, rough layout or sketch into professional slides. It is the presentation layer; your reviewed analysis remains the source of truth.
For the next step, explore the financial modelling agent pack and the Claude skills for pitch books.
Frequently asked questions
What is included in the download?
The complete pack contains 3 Claude skills with supporting instructions and examples. The directory below also offers one ZIP per skill so you can start with a single workflow.
How do I use it?
Download a single skill ZIP and upload it through the skill settings available in your Claude account. For Claude Code, keep each skill folder intact under the project’s .claude/skills/ directory. You can also add selected Markdown instructions to a Claude Project as reference material.
How does Oria fit?
Use Claude to research, test and structure the answer. Once you have reviewed the content, Oria helps turn the brief, rough layout or sketch into professional, editable slides.
Free download. Choose an individual skill below if you prefer to start with one workflow.