peth-eth

Intel Sweep Setup — Data skill for Claude Code

Data community

Claude Code skill: autonomous monitoring pipeline setup for deployed apps — signals, intel-digest endpoint, admin dashboard, and automation.

How to install Intel Sweep Setup

This entry records only its repository, not the path inside it, so there is no exact command to give. Open peth-eth/intel-sweep-setup and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Intel Sweep Setup does

Claude Code skill: autonomous monitoring pipeline setup for deployed apps — signals, intel-digest endpoint, admin dashboard, and automation.

Alternatives in Data

Claude Code Usage Monitor — by Maciek-roboblog - A real-time terminal-based tool for monitoring Claude Code token usage 7k ★
LLM Redteam — LLM red-team corpus runner — fires categorized prompt-injection / jailbreak / system-prompt-leak / data-exfil 4.5k ★
Claude Code Memory Setup — Up to 71.5x fewer tokens per session on Claude Code with Obsidian + Graphify 947 ★

README

intel-sweep-setup

A Claude Code skill that sets up autonomous monitoring for deployed apps — signals, intel-digest endpoint, admin dashboard, and automated pipeline. Works on any stack and host.

What it does

Gives you a continuous feed of crashes, user frustration, feature requests, and churn signals — without relying on users to report anything.

**End state:** A queryable `/api/intel-digest` endpoint that aggregates all intel from your app, an admin page to view it on demand, and an automated process that checks it regularly and surfaces findings wherever you work.

Two modes

Audit mode — `/intel-sweep-setup`

Scans what a project currently captures across three signal tiers, reports gaps, and offers to fix them.

Wire mode — `/intel-sweep-setup wire [project-name]`

Full setup: signals → endpoint → admin page → automation.

Signal tiers

Tier 1: Eyes Open (every app, day 1)

User counts (total, new today, DAU/WAU)
Client errors (crashes you'll never hear about otherwise)
Churn signals (paying users silent 3+ days)

Tier 2: Understanding Users (feature-dependent)

Onboarding funnel drop-off
Feature usage distribution
Chat topics, complaints, and feature wishes
Payment failures and cancellations
Trial-to-paid conversion
User feedback / PMF signals

Tier 3: Deep Ops (at scale)

Behavioral patterns (rage quit, burst-then-silence)
AI quality signals (regenerations, short responses)
Push notification health
Search dead ends
Rate limit hits
Cron / content freshness
Anti-abuse / bot detection

Farcaster add-on

Frame → app conversion, cast-to-engagement, social-graph churn.

Stack support

Detects your stack automatically:

Layer	Supported
DB	Prisma, Turso/libsql, Supabase, Drizzle, Mongoose
Platform	Vercel, Cloudflare Workers, Fly.io, Railway, VPS
Billing	Stripe, Lemon Squeezy
Observability	Sentry, PostHog, Umami, Vercel Analytics