# Lp Insights

> Generate hypothesis-driven landing-page feedback with simulated personas.

- Skill: `re-sianturi/lp-insights-2` (Agent Skill)
- Install (CLI): `npx skillmds@latest add re-sianturi/lp-insights-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/re-sianturi/lp-insights-2/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- License: MIT
- Author: re-sianturi (https://skillmd.com/u/re-sianturi)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/re-sianturi/lp-insights-2

---


# Landing Page Insights (OpenClaw Adapter)

OpenClaw project skill adapter for landing page persona analysis.

## Overview

Adapts the LP Insights methodology to the OpenClaw skill schema format.

## Usage

`/lp-insights <url> [traffic_source] [group_persona]`

## Core Loading & Security Procedure

Before starting the analysis, the agent must perform the following:
1. Load URL restriction lists and prompt injection rules from `core/SECURITY.md`.
2. Load and follow the execution rules in `core/WORKFLOW.md`.
3. Load the persona groups from `core/personas/_registry.md` and prompts from `core/prompts/persona-analysis.md`.
4. Validate outputs using schemas under `core/schemas/`.

## Runtime Policy & Tool Availability

The availability of browser tools and subagent execution is determined by the system's runtime security policy. The agent must check the runtime environment capabilities:
- **Capability Fallback:** If browser automation tools are unavailable, fallback to retrieving page source or text via standard text fetch tools. If subagents are unavailable, run simulation loops sequentially in the same execution context.
- **Tools:** Use standard system tools, never invent tool names.

## Methodology & Disclaimer

- Output declares `evidence_type` as `synthetic_hypothesis` for all analytics.
- Findings report exact count of affected personas, never claiming empirical conversion metrics or observed user behavior.