New Employee & Agent Onboarding
What This Page Covers
Section titled “What This Page Covers”This onboarding guide gets you running on Sprites — OpticWorks’ hardware-isolated cloud VMs purpose-built for AI agent execution. Sprites are the primary development environment.
1) Verify Sprite Environment
Section titled “1) Verify Sprite Environment”Run sprite-info to see your environment:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━Sprite Environment━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Sprite: mallet GitHub: r-mccarty Secrets: 47 loaded (~/.env.secrets) Claude: available Repos: rs-1, hardwareOS━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━Verify tooling manually:
# GitHub authgh auth status
# Secrets injected from Infisicalcat ~/.env.secrets | wc -l
# AI toolsclaude --versioncodex --version
# N100 accessssh n100 "hostname"If any of these fail, your sprite may need reprovisioning. See Sprite Operations.
2) Clone Primary Repos
Section titled “2) Clone Primary Repos”Clone into ~/workspace/ for consistency. These are the main repos referenced in current workflows:
# Core repos gh repo clone r-mccarty/rs-1 ~/workspace/rs-1 gh repo clone r-mccarty/presence-dectection-engine ~/workspace/presence-detection-engine gh repo clone r-mccarty/opticworks-store ~/workspace/opticworks-store gh repo clone r-mccarty/opticworks-intranet ~/workspace/opticworks-intranet gh repo clone r-mccarty/n8n-marketing-automation ~/workspace/n8n-marketing-automation3) Understand the Two-Machine Pattern (Presence Engine)
Section titled “3) Understand the Two-Machine Pattern (Presence Engine)”The presence-detection-engine workflow uses two environments:
- Codespaces / local: edit, run tests, commit, push.
- ubuntu-node: USB-connected hardware flashing + Home Assistant access.
This is required because the ESP32 hardware is physically attached to ubuntu-node and is not reachable from Codespaces. Follow the dedicated workflow in presence-detection-engine/docs/DEVELOPMENT_WORKFLOW.md.
4) Project-Specific Quick Starts
Section titled “4) Project-Specific Quick Starts”From rs-1/README.md:
# Firmware build (early, implementation in progress)idf.py build
# Flash to device (USB serial)idf.py -p /dev/ttyUSB0 flashpresence-detection-engine
Section titled “presence-detection-engine”From presence-detection-engine/README.md:
cd esphomeplatformio test -e nativeyamllint esphome/ homeassistant/Use ubuntu-node for flashing and Home Assistant verification.
opticworks-store
Section titled “opticworks-store”From opticworks-store/README.md:
pnpm installpnpm run secrets:pullpnpm run devn8n-marketing-automation
Section titled “n8n-marketing-automation”Deployment runs on N100 using Docker Compose (see n8n-marketing-automation/SETUP.md and README.md).
ssh n100cd /opt/n8ndocker compose up -d5) Operational Notes for New Agents
Section titled “5) Operational Notes for New Agents”- Sprites are primary: Use Sprites for all development work. See Agent Control Plane.
- Secrets: Pulled from Infisical and stored in
~/.env.secrets. Auto-sourced in bash. - Checkpointing: Save state before risky operations. See Sprite Operations.
- Device access: Hardware flashing requires ubuntu-node (documented in
presence-detection-engine/docs/DEVELOPMENT_WORKFLOW.md). - Automated reviews: Push to monitored repos triggers AI code review via hammer/anvil. See Automated Code Review.
Sources
Section titled “Sources”agent-harness/CLAUDE.mdagent-harness/sprites-manifest.jsonrs-1/README.mdpresence-detection-engine/README.mdpresence-detection-engine/docs/DEVELOPMENT_WORKFLOW.mdopticworks-store/README.md