mercury-heartbeat

Run WATCH, WORK, LEARN, and DREAM maintenance cycles on repositories.

Updated Jul 7, 2026
One-click install
npx skills add https://github.com/TitoPrausee/nexus-toti --skill mercury-heartbeat
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mercury-heartbeat
Source: https://github.com/TitoPrausee/nexus-toti/tree/main/data/skills/productivity/mercury-heartbeat
Command: npx skills add https://github.com/TitoPrausee/nexus-toti --skill mercury-heartbeat

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates recurring project maintenance so an AI agent can monitor health, execute prioritized work, learn from patterns, and generate improvement ideas without manual orchestration.

Core Features & Use Cases

  • WATCH phase: Checks RAM, git status, tests, and essential services before any action is taken.
  • WORK phase: Pulls the highest-priority backlog item, detects whether it is already implemented, and completes one task per cycle.
  • LEARN and DREAM phases: Reviews near-term work, builds micro-prototypes, captures notes, and records gaps, risks, shortcuts, and ideas for future cycles.
  • Use Case: Run a scheduled heartbeat on a long-lived codebase to keep maintenance moving, surface issues early, and preserve a durable history of actions and lessons.

Quick Start

Use the mercury-heartbeat skill to run a full WATCH WORK LEARN DREAM cycle for the current project and report the results.

Frequently Asked Questions about mercury-heartbeat

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate project maintenance for an autonomous git repository?

Automated project maintenance for an autonomous git repository runs scheduled health checks, executes prioritized backlog items, and records improvement ideas without manual orchestration. It applies phased orchestration to monitor RAM, inspect git state, and run tests continuously.

What is a cron-driven health check for long-running codebases?

A cron-driven health check for long-running codebases is an automated cycle that verifies RAM usage, git status, and test execution before pulling backlog tasks. It ensures continuous project maintenance and surfaces issues early through durable cycle reporting.

Can an autonomous agent detect if a backlog item is already implemented?

An autonomous agent can detect if a backlog item is already implemented during its work phase by inspecting the current git state. It pulls the highest-priority task, checks existing implementations, and completes one task per cycle.

How do I set up continuous self-improvement cycles for software engineering workflows?

Continuous self-improvement cycles for software engineering workflows are set up by scheduling a heartbeat that executes WATCH, WORK, LEARN, and DREAM phases. This reviews near-term work, builds micro-prototypes, and records gaps, risks, and ideas for future cycles.

Does autonomous project maintenance require external dependencies to monitor RAM and git status?

Autonomous project maintenance requires no external dependencies to monitor RAM and git status. It relies entirely on reliable internal git state inspection, test execution, and RAM monitoring to execute phased orchestration and durable cycle reporting.

When should I not use a heartbeat cycle for project maintenance?

You should not use a heartbeat cycle for project maintenance if your workflow lacks reliable git state inspection or test execution capabilities. The phased orchestration requires a stable environment to accurately pull backlog items and monitor RAM without errors.