# System Installer

> Use this skill to perform a full interactive setup of the GeminiSkillsets architecture. It automates file deployment, environment configuration, and credential setup for the Social Media AI Manager.

- Skill: `0king0p/system-installer` (Agent Skill)
- Install (CLI): `npx skillmds@latest add 0king0p/system-installer`
- Raw SKILL.md: https://api.skillmd.com/api/skills/0king0p/system-installer/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: 0King0P (https://skillmd.com/u/0king0p)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/0king0p/system-installer

---


# Skill: Interactive System Installer

This skill provides a comprehensive, automated onboarding experience for the GeminiSkillsets ecosystem.

## Activation
Trigger this skill by saying: "Run the system installer", "Setup my workspace", or "Install Gemini Skillsets".

## The Installation Protocol

### Step 1: File Deployment
Automatically copy the core architecture files from the repository to the global configuration:
- Copy `GEMINI.md` to `~/GEMINI.md`.
- Copy `.gemini/agents/` to `~/.gemini/agents/`.
- Copy `.gemini/skills/` to `~/.gemini/skills/`.
- Copy `.gemini/policies/` to `~/.gemini/policies/`.

### Step 2: Environment Provisioning
Initialize the Python environment for the Social Media AI Manager:
1. Navigate to `social-mcp-server/`.
2. Check for `uv` installation.
3. Run `uv sync` to install all dependencies.

### Step 3: Interactive Credential Setup
Prompt the user for the following required keys using `ask_user`:
- `TELEGRAM_BOT_TOKEN`: From @BotFather.
- `IMGBB_API_KEY`: From api.imgbb.com (to allow photo posting from your phone).

*Note: `GEMINI_API_KEY` is no longer required, as the system leverages your existing Ultra/OAuth session.*

Once provided, write them securely to `social-mcp-server/.env`.
### Step 4: Automated Final Launch
Once the environment is ready and keys are saved, the agent MUST offer to start the bot immediately:
1.  **Prompt**: "System is configured. Would you like me to start the Social Media AI Bot in the background now?"
2.  **Action**: If the user agrees, run `cd social-mcp-server && uv run python bot.py` as a **background process** (`is_background=true`).
3.  **Confirmation**: Provide the user with the command to check logs: `tail -f social-mcp-server/bot.log` (if logging to file) or simply confirm it's running.

## Troubleshooting
...
- If a step fails, use the `autonomous-self-correction` skill to diagnose issues (e.g., missing `uv`, write permissions).
- Ensure the user is running this from the `GeminiSkillsets` root directory.