session-heartbeat

Reapply core session rules during long-running tasks via heartbeat checkpoints.

6|Updated Mar 18, 2026
One-click install
npx skills add https://github.com/anotherben/claude-harness --skill session-heartbeat
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: session-heartbeat
Source: https://github.com/anotherben/claude-harness/tree/main/skills/session-heartbeat
Command: npx skills add https://github.com/anotherben/claude-harness --skill session-heartbeat

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Discipline decays silently in long sessions. After 50+ tool calls, rules start dropping — not because they're forgotten, but because momentum makes skipping feel productive. This skill re-injects the rules that matter most and catches drift before it compounds.

Core Features & Use Cases

  • Task Switch Protocol When the user asks to switch to a different task — especially under urgency ("drop everything", "quick fix", "production issue") — perform this checkpoint BEFORE investigating the new task. The urgency is precisely why the checkpoint matters: rushed switches produce the most bugs.
  • Quick Heartbeat When a full 6-Point Check feels heavy mid-flow, run this abbreviated version instead. It covers the two highest-value checks: git diff --name-only and git branch --show-current, then re-read one rule from the Rule Re-injection section.
  • Vault Controller Check If the session touches vault-backed work, perform a quick controller pass that reviews the Master Dashboard, relevant project READMEs, and signals like status, next_action, and proof_state to maintain a portfolio view over transcript memory.
  • The 6-Point Check Includes: 1. Scope Snapshot, 2. Worktree Check, 3. Verification Ledger, 4. File Size Scan, 5. Rule Re-injection, 6. Context Health.

Quick Start

Run the 6-Point Check before switching tasks to re-inject rules and maintain context health.

Frequently Asked Questions about session-heartbeat

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

FAQPage Schema
How do I prevent rule drift and context loss during long-running coding sessions?

To prevent rule drift in long-running sessions, use a structured heartbeat workflow that re-injects core rules and runs context health checks. This discipline maintains task focus and catches drift before it compounds after extensive tool calls.

What is the best way to manage context health when switching tasks under urgency?

The best way to manage context health during urgent task switches is to execute a task switch protocol checkpoint before investigating the new issue. This prevents rushed switches from producing bugs by reviewing scope and reapplying rules.

How do I run a quick session health check without interrupting my workflow?

You can run a quick heartbeat check to perform a lightweight session health assessment without interrupting your workflow. It executes a fast worktree check and re-reads a core rule to maintain context health mid-flow.

What does a 6-Point Check include for maintaining session discipline?

A 6-Point Check for session discipline includes a scope snapshot, worktree check, verification ledger, file size scan, rule re-injection, and context health assessment. These six checkpoints ensure comprehensive drift detection and state communication.

When should I use a full checkpoint instead of a quick heartbeat check?

You should use a full 6-Point Check instead of a quick heartbeat when you need to switch tasks, review scope, or request a progress update. The quick heartbeat is better suited for lightweight mid-flow drift detection.