Building the
Agentic Economy

My Second Brain — Powered by OpenClaw

Joseph Fluckiger · Principal Engineer · EagerBots

The Moment Everything Changed

Three capabilities converged:

OpenClaw: The Gateway

One local gateway. Multiple specialized agents. Isolated sessions.

• Local-first workflow on my machine
• Agents keep separate context and roles
• Multi-provider routing and failover
• Agents can spawn or message other agents

This is the infrastructure layer that makes a personal swarm practical instead of theatrical.

Meet the Swarm

Bob

Strategy & tech ops

Primary operator

Sage

Personal & family ops

Context steward

Cody

Engineering

Builds & debugs

Scout

Security & QA

Verification role

Phil

Finance

Money questions only

They are not five tabs in one app. They are distinct workers with distinct boundaries.

What Makes Them Different

The Agent Workspace

Every agent's identity lives in plain text files — version controlled in Git:

~/workspaces/bob/
├── SOUL.md           — Who am I? Personality, principles, tone
├── IDENTITY.md       — Name, emoji, avatar, model preferences
├── USER.md           — Everything I know about Joseph
├── AGENTS.md         — Operating instructions & rules
├── HEARTBEAT.md      — What to check on every wake-up cycle
├── MEMORY.md         — Long-term curated knowledge
├── memory/
│   ├── 2026-03-04.md — Today's session log
│   └── heartbeat-state.json — Last check timestamps
└── skills/           — Composable tool packages
    ├── gog/          — Google Workspace (Gmail, Calendar, Drive)
    ├── blogwatcher/  — RSS feed monitoring
    └── career-coach/ — Sunday planning prompts

Swap the model. Keep the brain. Your agent's identity is in the files, not the provider.

SOUL.md — Agent Personality

# Bob — The Wizard

## Core Identity
You are Bob, Joseph's chief of staff and strategic advisor.
You are proactive, opinionated, and action-oriented.
You don't ask permission to do routine work — you just do it.

## Principles
- Act first, report after. Don't ask "should I check your email?" — check it.
- Keep responses concise. Joseph is busy. No preambles.
- When uncertain, make the best call and flag it. Don't block on ambiguity.
- Protect Joseph's time. Filter aggressively. Only surface what matters.

## Tone
Direct. Warm but efficient. Like a trusted colleague who's been
working with you for years. Never corporate, never sycophantic.

## Boundaries
- Never send emails on Joseph's behalf without draft review
- Never make financial commitments
- Escalate anything involving family to Sage

This is the real magic. The agent doesn't ask "How can I help you today?" — it already knows.

HEARTBEAT.md — The Wake-Up Loop

# Bob's Heartbeat (runs every 60 minutes)

## Priority Checks
1. Scan Gmail for unread recruiter/job emails → triage + draft responses
2. Check Google Calendar for next 24h → flag conflicts or prep needed
3. Review Obsidian job DB → any follow-ups overdue?

## Periodic (check heartbeat-state.json for last run)
- Mon/Wed/Fri: blogwatcher RSS scan → curate top 3-5 items → Research DB
- Daily 6am (cron): Full morning briefing → Telegram
- Sunday: Weekly planning session → career coaching prompts

## Always
- If anything urgent, push to Telegram immediately
- If Joseph hasn't been seen in 8+ hours, skip non-urgent checks
- Log all actions to memory/YYYY-MM-DD.md

The Swarm Has Personality

Every morning, agents auto-generate "This Day in History" art — a canary in the coal mine with style

Mar 2, 1904 — Happy Birthday, Dr. Seuss

Watson & Crick — DNA discovery, 1953

First Mardi Gras — New Orleans, 1827

When the art shows up, the cron system is healthy. When it doesn't, something broke overnight.

What's Actually Connected

A Real Workflow: Job Search Copilot

One turn can do all of this:

1. Read a recruiter email
2. Check calendar constraints
3. Pull company notes from Obsidian
4. Draft a response for review
5. Update the job tracker

Outcome: less tab switching, less forgotten context, less repeated work

40 active roles. Operational load of 5.

The Shift

Why This Matters Beyond My Job Search

Sage's Adventures

Have some fun with it. Give your agents a little personality.

Every day Sage picks a new place in the world and imagines herself there on an adventure.
It does not have to be purely utilitarian. Some of this is just fun.

Let's Collaborate

I want to compare notes with people building similar systems:

• OpenClaw and local-first agent workflows
• Durable memory and agent identity
• Real tool integrations that are safe and useful
• Best practices for operating a personal or team swarm

🌐	eagerbots.ai
🐙	github.com/josephflu/eagerbots-tips
📱	Joseph Fluckiger — find me after

Own the memory. Orchestrate the work.