s10-eternal-self-heal

Runs daily audit prompts to detect, remediate, and verify system health drift.

Updated Jun 20, 2026
One-click install
npx skills add https://github.com/Noetfield-Systems/SourceA --skill s10-eternal-self-heal-noetfield-systems
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: s10-eternal-self-heal
Source: https://github.com/Noetfield-Systems/SourceA/tree/main/agent-skills/shared/s10-eternal-self-heal
Command: npx skills add https://github.com/Noetfield-Systems/SourceA --skill s10-eternal-self-heal-noetfield-systems

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Long-running agent systems drift over time: monitors go stale, inbox state diverges from disk truth, validators weaken, and skills fall out of sync. This Skill provides a scheduled self-healing audit loop that continuously detects, classifies, remediates, and verifies these failures before they compound. ## Core Features & Use Cases - Daily audit rotation: Executes 10 audit prompts per day from a 100-prompt manifest across 10 domain packs (disk truth, self-heal, E2E, skills sync, validators, governance, and more), with a full 100-prompt run every Sunday UTC. - Disk-truth self-heal: Automatically remediates script-owned components such as monitor live sync, run inbox disk truth, skills sync, and the phase-strict queue, while routing hub UI and freeze-button issues to maintainer review. - Receipts and gating: Writes machine-readable receipts and history to disk, with a validator gate that blocks claims of "system fixed" while any S10 check fails. - Use Case: After an incident where the monitor stopped scrolling, run the daily S10 loop to detect the drift, auto-heal the monitor sync, and produce a receipt proving the system passed validation. ## Quick Start Ask the agent to run today's S10 eternal audit and report any failing checks with their remediation status.

Frequently Asked Questions about s10-eternal-self-heal

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

FAQPage Schema
How do I run a daily system health audit for an agent system?

Run the daily wrapper command (~/.sina/bin/s10-eternal-daily) from any directory to execute today's 10 audit prompts. The pack rotates by day of year, and results are written to a receipt file with a validation gate.

How do I audit all 100 self-heal prompts at once?

Run the full audit with ~/.sina/bin/s10-eternal-full or the script's --full flag. This executes all 100 prompts across the 10 packs and is scheduled weekly on Sunday UTC.

What does the self-heal loop auto-remediate versus escalate?

It auto-heals script-owned components: monitor live sync, run inbox disk truth, skills sync, phase-strict queue, and governance history. Hub UI, freeze buttons, and dual-pick issues are not auto-healed and go to maintainer review.

Can I claim the system is fixed if S10 checks still fail?

No. The validator gate treats any failing S10 check as proof the system is not fixed. You must remediate failures and pass the validation script before declaring recovery.

Why is the S10 audit receipt considered stale?

A receipt older than 26 hours triggers a warning at session start because the daily loop has not run. Re-run the daily audit to refresh the receipt and restore a current health baseline.