# telnyx-rag

> Semantic search over workspace files using Telnyx Storage + AI embeddings. Index your memory, knowledge, and skills for natural language retrieval.

- Author: Chief Bot
- Repository: dcasem/clawdbot-recipes
- Version: 20260127075829
- Stars: 2
- Forks: 0
- Last Updated: 2026-02-06
- Source: https://github.com/dcasem/clawdbot-recipes
- Web: https://mule.run/skillshub/@@dcasem/clawdbot-recipes~telnyx-rag:20260127075829

---

---
name: telnyx-rag
description: Semantic search over workspace files using Telnyx Storage + AI embeddings. Index your memory, knowledge, and skills for natural language retrieval.
---

# Telnyx RAG Memory

Semantic search over your Clawdbot workspace using Telnyx's native embedding and similarity search APIs.

## Requirements

- **Telnyx API Key** — that's it!
- Get one free at [portal.telnyx.com](https://portal.telnyx.com/#/app/api-keys)

## Quick Start

```bash
# Set your API key
export TELNYX_API_KEY="KEY..."

# Run setup
cd ~/skills/telnyx-rag
./setup.sh

# Search your memory
./search.py "What are my preferences?"
```

## What It Does

- **Indexes** your workspace files (MEMORY.md, memory/*.md, knowledge/, skills/)
- **Embeds** content automatically using Telnyx AI
- **Searches** using natural language queries
- **Prioritizes** results from memory/ (your primary context)

## Setup Options

### Option 1: Environment Variable
```bash
export TELNYX_API_KEY="KEY..."
./setup.sh
```

### Option 2: .env File
```bash
echo 'TELNYX_API_KEY=KEY...' > .env
./setup.sh
```

### Custom Bucket Name
```bash
./setup.sh my-custom-bucket
```

## Usage

### Search Memory

```bash
# Basic search
./search.py "What are David's communication preferences?"

# More results
./search.py "meeting notes" --num 10

# JSON output (for scripts)
./search.py "procedures" --json
```

### Sync Files

```bash
# One-time sync
./sync.py

# Check status
./sync.py --status

# List indexed files
./sync.py --list
```

## Configuration

Edit `config.json` to customize what gets indexed:

```json
{
  "bucket": "clawdbot-memory",
  "region": "us-central-1",
  "workspace": ".",
  "patterns": [
    "MEMORY.md",
    "memory/*.md",
    "knowledge/*.json",
    "skills/*/SKILL.md"
  ],
  "priority_prefixes": ["memory/", "MEMORY.md"]
}
```

## How It Works

```
┌─────────────────┐     ┌─────────────────────────────┐
│  Your Workspace │     │     Telnyx Cloud            │
│  ├── memory/    │────▶│  Storage: your-bucket/      │
│  ├── knowledge/ │     │     └── (your files)        │
│  └── skills/    │     │              │              │
└─────────────────┘     │              ▼ auto-embed   │
                        │     Telnyx AI Embeddings    │
        Query ─────────▶│              │              │
                        │              ▼              │
        Results ◀───────│     Similarity Search       │
                        └─────────────────────────────┘
```

## Clawdbot Integration

Add to your `TOOLS.md`:

```markdown
## Semantic Memory

Search memory semantically:
\`\`\`bash
cd ~/skills/telnyx-rag && ./search.py "your query"
\`\`\`
```

### Automated Sync

Add to your heartbeat or cron:
```bash
cd ~/skills/telnyx-rag && ./sync.py --quiet
```

## Troubleshooting

### "Bucket not found"
```bash
./sync.py --create-bucket
```

### "No results found"
- Wait 1-2 minutes after initial sync (embeddings take time)
- Check files are uploaded: `./sync.py --list`

### Credential errors
- Verify `TELNYX_API_KEY` is set: `echo $TELNYX_API_KEY`
- Or check `.env` file exists in skill directory

## API Reference

### From Python

```python
from search import similarity_search

results = similarity_search("What do I know about X?", num_docs=5)
for r in results:
    print(r["source"], r["content"])
```

## Credits

Built for [Clawdbot](https://github.com/clawdbot/clawdbot) using [Telnyx Storage](https://telnyx.com/products/cloud-storage) and AI APIs.