Hey team, I made a system that is meant to produce automated daily / weekly / monthly / yearly regen digests. It’s also meant to host regenai docs. This is very early stage, just got the concept up. I plan on making improvements, and am open to ideas and feedback.
Here’s the roadmap:
# Regen Heartbeat — Three-Phase Roadmap
This document describes the three phases of the Regen Heartbeat project. Each phase
builds on the previous one, delivering working functionality at every stage. For full
architectural context, see `@.claude/specs/regen-heartbeat.md`.
---
## Phase 1 — Skeleton & Daily Heartbeat
**Goal:** Build the core infrastructure and deliver a working daily digest.
**What gets built:**
- Complete directory structure with README.md and CLAUDE.md in every directory
- `settings.json` at the repository root with template defaults, current events
topics, and site configuration
- The full character pipeline: 12 output styles derived from character JSONs,
12 character skills that load those output styles, and the routing logic that
connects `--character` flags to the appropriate skill
show original
And the spec:
# Regen Heartbeat — System Specification
This document is the single source of truth for the Regen Heartbeat project. Every
implementation decision traces back here. If something isn't in the spec, it isn't in
scope. If something in the spec is ambiguous, resolve it here before building.
---
## 1. What This Is
Regen Heartbeat is a dual-purpose system.
**First**, it is an automated digest engine. It generates structured reports about
Regen Network, RegenAI, and the Regen Commons at daily, weekly, monthly, and yearly
cadences. These digests are published as markdown files in a GitHub repository and
served as a static site via Quartz on GitHub Pages.
**Second**, it is an educational template. It demonstrates how to work with Claude Code
effectively — commands, skills, agents, MCPs, plugins, hooks, output styles, and
character systems. Every structural choice should be exemplary. A newcomer should be
show original
And a reflection on the construction methodology:
# Regen Heartbeat: Case Study Report
## Building an Automated Ecosystem Observatory with Claude Code and Live MCP Data
*February 3, 2026 — First Production Run*
---
## 1. What Was Built
Regen Heartbeat is an automated digest engine that pulls live data from the Regen
Network blockchain, the KOI knowledge commons, and the broader web, then synthesizes
it into structured reports published as a static site via Quartz on GitHub Pages. It
also functions as a teaching repository for Claude Code — every structural choice is
designed to be studied and learned from.
Today, the system generated its first daily digest. The report covered on-chain
governance (Proposal #62, the v7.0 upgrade), ecocredit activity (13 credit classes,
78 batches, 58 projects), chain health (224.76M REGEN supply, 21 validators),
ecosystem intelligence (Tokenomics WG, Anti-Krisis framework, Regen AI launch), and
show original
Discussions with @brawlaphant and @Gregory_Regen inspired this build. And thanks to @darrenZ for building the Regen KOI MCP that powers this and @Jeancarlo for building the original Regen MCP that the Regen Python MCP is based off of.
2 Likes