# Vector DB Init

> Interactively initializes the Vector DB plugin. Installs the required pip dependencies (chromadb, langchain wrappers) and configures the vector_profiles.json for Native Python Server connections. Run this before attempting to use the vector-db-search or vector-db-launch skills.

- Skill: `richfrem/vector-db-init-3` (Agent Skill, multi-file: 9 files)
- Install (CLI): `npx skillmds@latest add richfrem/vector-db-init-3`
- Raw SKILL.md: https://api.skillmd.com/api/skills/richfrem/vector-db-init-3/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: richfrem (https://skillmd.com/u/richfrem)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/richfrem/vector-db-init-3

---

# Vector DB Initialization

The `vector-db-init` skill is an automated setup routine that prepares the environment for the Vector DBMS. 

## Examples

Real-world examples of each config file are in `references/examples/`:

| File | Purpose |
|:-----|:--------|
| [`vector_profiles.json`](../../references/examples/vector_profiles.json) | Profile registry -- defines named vector collections and ChromaDB connection |
| [`vector_knowledge_manifest.json`](../../references/examples/vector_knowledge_manifest.json) | Manifest -- what folders/globs to include/exclude in the vector index |

## When to Use This
- When a user first installs the `vector-db` plugin.
- If the user complains that `chromadb` is not installed or `ModuleNotFoundError` is thrown.
- If the Vector DB profile is missing from `.agent/learning/vector_profiles.json`.

## Instructions for Agent

1. **Run the Initialization Script:**
   You must execute the interactive initialization script located at `scripts/init.py`.
   ```bash
   python3 ./scripts/init.py
   ```

2. **Wait for Completion:** 
   The script will automatically run `pip install`, then prompt the user to select their deployment architecture (In-Process or Native Server). All settings are written to `.agent/learning/vector_profiles.json`.

3. **Verify:**
   After the script completes successfully, inform the user that their environment is ready, and they can now run the `vector-db-launch` skill to start the background server (if they chose Option 2).