harryvondiesel-web /
5-persona-advisory-board
A prompt file that makes an AI interview you, then judge a big decision five ways
- 136 stars
- 8 stars since 3 Oct
- Updated 10 days ago
- Beginner friendly
- Agent skills & plugins
Most AI advice sounds sure of itself before it knows your situation. This skill tries to fix that. When you give it a decision, it checks whether it has enough context. If not, it interviews you instead of guessing.
Once it has enough, it applies five separate lenses to the same facts: Product Clarity, Risk and Capital, Scale and Systems, Offer Strength, and Future Self. The first four are inspired by public principles linked to Jobs, Buffett, Bezos and Hormozi. The README says they are analytical lenses, not simulations of those people. The workflow is called CRIT: Context, Role, Interview, Task.
The output is structured: a context read, the five views, where they disagree, a future-self check, and a final brief. It is a prompt, not software. The README says it reorganizes the context you give it and does not prove market demand.
What it does
Interview before analysis
If the decision lacks context, the skill asks targeted questions first. Pricing, hiring and stop/continue decisions each get different questions.
Five distinct lenses
Product Clarity, Risk and Capital, Scale and Systems, Offer Strength, and Future Self each look at the same evidence from a different angle.
Surfaces disagreement
The output shows where the lenses conflict rather than blending them into one bland answer.
Structured decision brief
Results come as Decision, Context Read, 5 Lens Views, Disagreement, Future Self Check and a Final Decision Brief with one next action.
Works across platforms
Install paths cover Claude Projects, ChatGPT Projects or Custom GPTs, Codex, Hermes and OpenClaw. The README marks ChatGPT as expected rather than tested.
What you could build
- Decide whether to launch an offer now or keep refining it.
- Test a price increase before announcing it to customers.
- Weigh whether to accept or turn down a risky client.
- Judge whether to build a new feature or stay focused.
- Decide whether to stop, continue or restructure a struggling project.
Get started
For Claude or ChatGPT, create a Project and paste the full contents of SKILL.md into its instructions, then start a chat and describe your decision. For Codex, copy SKILL.md into the skills folder with the commands below and start a fresh session.
mkdir -p ~/.codex/skills/5-persona-advisory-boardcp SKILL.md ~/.codex/skills/5-persona-advisory-board/SKILL.mdCommands are copied from the project's README. Read it on GitHub for the full setup.
More in Agent skills & plugins
superpowers
A set of skills that makes coding agents plan, test and review before they write code
mattpocock/skills
Small, editable agent skills for planning, testing and reviewing code with AI
ECC
Agents, skills, hooks and memory that add an engineering workflow to coding agents
andrej-karpathy-skills
A CLAUDE.md file that tells Claude Code to ask first, keep code small, edit narrowly
anthropics/skills
Anthropic's example Agent Skills: folders of instructions Claude loads per task
ponytail
A skill that makes AI coding agents write the least code that works
gstack
Slash-command skills that give Claude Code roles like reviewer, QA and release engineer
addyosmani/agent-skills
A pack of 25 Markdown skills that guide AI coding agents from spec to ship
taste-skill
Agent skills that steer AI coding tools toward better layout, type and motion
archify
Agent skill that turns a description or a codebase into an interactive HTML diagram