Peoples — Development skill for Claude Code

Peoples

Peoples is a market-intelligence and persona-simulation skill for [Claude Code](https://code.claude.com). It is prompt-only: one `SKILL.md`, with no scripts and no hooks. When a product, module or user-facing feature is proposed, it makes the model look at competing products, find what their users complain about, fix those flaws in the current code, and then run a synthetic persona panel that estimates adoption and lists the main objections.

Rules

Section	Rule	What the model does
1	Identify competitors	It names 3–5 existing market alternatives or similar open-source or commercial products.
1	Mine negative feedback	It collects common complaints, bad reviews and UX friction points of those competitors (poor error handling, slow response, complex onboarding, missing edge cases).
1	Preemptive patching	It fixes those mined flaws in the current code or architecture before delivering the product.
2	Persona panel	It evaluates the proposal with 100–1000 synthetic personas from five groups: Tech Savvy, Beginner, Enterprise, Budget-Conscious and Security-Obsessed.
2	Scoring	Each persona gives a usability and friction score (1–10), a value proposition match (1–10) and its deal-breakers.
2	Report	It emits the PEOPLES SIMULATION REPORT: flaws patched in code, adopt and reject percentages, top rejection reasons, and the code change needed for 90 %+ adoption.
3	Background execution	Research and persona testing run in an isolated block or parallel thread so code generation is not delayed. Only aggregated metrics are shown, not per-agent logs.
4	Pre-flight checklist	Before delivery it checks that competitor flaws were patched, the cohort was run, and percentages and objections were calculated.

Install

git clone https://github.com/omer-faruk-g/Peoples ~/.claude/skills/peoples

Then restart Claude Code. The skill loads when a product or user-facing feature is propose