Run product work as voice note → grill-the-idea → Linear spec → parallel agent build → persona browser QA → Sentry fix PRs, so humans only see version five: AI agent prompt
On Nov 3, 2026 Greg Isenberg posted a widely shared X thread (hundreds of thousands of views) about running a 50+ person, 12-country company by talking to agents more than teammates. The stack: voice note to Grok Bot; mattpocock/skills /grill-me to interrogate the idea; Linear MCP for the spec; Exa/Firecrawl research plus Apify review mining for top user complaints; four design options before code; parallel agents in git worktrees; headless persona testing on Vercel previews; Sentry MCP opening fix PRs in under 10 minutes; PostHog cron watching funnel drops. Operating rules: one agent per platform, cheap models for triage and Opus for shipped work, LLM judge scores under 8 get sent back, mistakes become reusable skills. Humans get version five, not the raw voice note.
The prompt
Help me install a Greg-Isenberg-style agent product loop for [my product]. Stack I have or will use: [Grok Bot / Claude / Cursor], Linear, [Exa/Firecrawl], git worktrees, Vercel, Sentry, PostHog. Hard rules: agents never ship to production without my approve; no password in browser agents—I take over at login; one agent per platform; score every output 1–10 and redo under 8. 1) Map my workflow stages and which agent owns each. 2) Draft the /grill-me and research prompts (top 10 quoted complaints). 3) Spec overnight build + morning review checklist. 4) Define the Sentry→fix-PR and PostHog funnel-drop cron prompts. Output a one-page operating manual I can paste into each agent.
Fill in the blanks, then copy
This prompt has 3 [bracket]s to fill in. Type your details below and the prompt updates as you go. Your answers stay in this browser and are reused on other prompts with the same blanks. Nothing is sent to us.
Example values are fictional, just to show how a filled-in prompt reads. Replace them with your real details.How to use it
- Copy the prompt above.
- Replace every [bracket] with your own details (use the fill-in-the-blanks box above, which remembers your answers for the next prompt).
- Paste it into Grok Bot, Claude/Opus, Linear MCP and review the answer before you use it.
Works with: Grok Bot, Claude/Opus, Linear MCP, Exa, Firecrawl, Apify, Vercel, Sentry MCP, PostHog MCP, Cursor
Real-world example: x.com/gregisenberg/status/2117734977312047537
Before you use it: what to check
Already built into this prompt
- Hard rules: agents never ship to production without my approve; no password in browser agents—I take over at login; one agent per platform; score every output 1–10 and redo under 8.
Check yourself
- Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details.
Using an AI agent? Download this prompt as a SKILL.md agent skill with these checks built in, or get the .zip for Claude.ai. How to install skills.
Want to make this prompt even better? Grade it with the free prompt grader or build your own with the AI prompt generator.
32 copy-paste agent prompts (PDF) plus 5 new prompts every week.
Get it free