publish-and-reuse

Synchronizes Personal AI infrastructure across devices using git ancestry-based direction resolution.

3|Updated Feb 13, 2026
One-click install
npx skills add https://github.com/ooooooooooooooooooop/personal-ai --skill publish-and-reuse-ooooooooooooooooooop
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: publish-and-reuse
Source: https://github.com/ooooooooooooooooooop/personal-ai/tree/main/skills/publish-and-reuse
Command: npx skills add https://github.com/ooooooooooooooooooop/personal-ai --skill publish-and-reuse-ooooooooooooooooooop

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Keeping a multi-device Personal AI setup consistent is error-prone: users must manually decide whether to pull, push, merge, or restore across public repos, private state repos, project repositories, and runtime configs. This Skill removes that burden by automatically classifying each repository's sync state and executing only provably safe actions. ## Core Features & Use Cases - Automatic direction resolution: One command classifies every plane (personal-ai, personal-ai-private, projects, memory, runtime) as PULL, PUSH, MERGE, NO ACTION, REVIEW, or BLOCKED using git ancestry rather than timestamps. - Deterministic memory merge: Merges Dynamic Memory records and immutable revisions across devices via the MemoryProvider contract, flagging concurrent revisions instead of overwriting them. - Safe restore and runtime refresh: Performs fresh-device restore (clone, bootstrap, render, verify) and regenerates only affected Harness configs with snapshot, post-diff, and automatic rollback. - Use Case: You sit down at a second computer and say "sync my Personal AI" — the Skill fetches all remotes, fast-forwards clean repos, merges new memory records, refreshes affected runtime configs, and reports a short PASS/REVIEW/BLOCKED summary. ## Quick Start Ask the agent to sync my Personal AI and automatically determine whether to pull, push, merge, or restore on this machine.

Frequently Asked Questions about publish-and-reuse

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

FAQPage Schema
How do I sync my AI environment across multiple computers?

Run the sync entry point, which fetches all remotes, classifies each repository by git ancestry, and executes only safe actions such as clean fast-forward pulls and deterministic memory merges. Dirty worktrees and diverged histories are reported for manual review instead of being touched.

How do I restore my Personal AI setup on a new machine?

Use the restore mode, which clones the canonical and private repositories, runs bootstrap and device discovery, renders installed Harness configs, rebuilds the memory index, and verifies session history. It is idempotent and skips secrets that are not required for the active route.

Does git sync handle secrets and API keys?

No, secrets never enter git synchronization. The sync only detects whether each secret reference is AVAILABLE, MISSING, or NOT_REQUIRED, and missing optional secrets do not block the overall sync.

What happens when both devices modified the same files?

Diverged infrastructure repositories are marked REVIEW and never auto-merged or force-pushed. The sole exception is disjoint Dynamic Memory record changes, which merge deterministically while concurrent revisions of the same record are preserved and flagged.

Why does sync report REVIEW instead of pushing my local commits?

Automatic sync never performs implicit commits or pushes. Local-ahead history is only pushed in explicit push mode when every commit carries a valid ownership receipt, validation gates pass, and a privacy scan finds no leaks.