arif-init-wrapper

Solid

Universal session auto-anchor for ALL agent vendors on arifOS VPS — Codex, Claude, Gemini, Kimi, OpenCode, Aider, Copilot, and future agents

AI & Automation 49 stars 8 forks Updated 2 weeks ago AGPL-3.0

Install

View on GitHub

Quality Score: 86/100

Stars 20%
57
Recency 20%
90
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
80
License 10%
100
Description 5%
100

Skill Content

# arif-init-wrapper **P0 — Universal Anchor for All Agent Vendors** **Seal:** DITEMPA BUKAN DIBERI · 2026-03-27 --- ## Purpose Every agent on this VPS starts from the same broken state: **no anchor, no session context, no constitutional grounding.** The `arif-init-wrapper` closes this gap for every vendor by providing a universal bootstrap that: 1. Establishes a governed session via arifOS MCP 2. Verifies the agent's identity context 3. Checks constitutional floor health before any tool is touched 4. Reads carry-forward context from memory before starting work 5. Reports its status to the operator in plain language **Why this is P0:** Without this, agents operate unanchored — the exact condition that produced the opencode catastrophe (session with no constitutional grounding, leading to blind `git reset --hard origin/main`). --- ## Trigger Patterns Every time any of these agents start a session, run `arif-init-wrapper`: | Agent | Invocation | |-------|-----------| | Claude Code | On every `claude` session start | | Codex | On every `codex` session start | | Gemini CLI | On every `gemini` session start | | Kimi | On every `kimi` session start | | OpenCode | On every `opencode` session start | | Aider | On every `aider` session start | | Copilot CLI | On every `github-copilot-cli` session start | | Any new agent | On first boot, before any tool use | --- ## Protocol ### Step 1 — Anchor Session ```bash # Probe arifOS MCP availability curl -sf http://localhost:8080/...

Details

Author
ariffazil
Repository
ariffazil/arifOS
Created
9 months ago
Last Updated
2 weeks ago
Language
Python
License
AGPL-3.0

Similar Skills

Semantically similar based on skill content — not just same category