directory-indexer 0.0.5 → 0.0.6

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -37,10 +37,15 @@ directory-indexer serve
37
37
 
38
38
  Now ask Claude: _"Find files similar to my Redis incident reports"_ and it will search your indexed documents semantically.
39
39
 
40
- ## Requirements
40
+ ## Setup
41
41
 
42
- - **Qdrant**: Vector database for semantic search
42
+ Before using directory-indexer, you need to set up two services:
43
43
 
44
+ ### 1. Qdrant Vector Database
45
+
46
+ Choose one option:
47
+
48
+ **Docker (recommended for most users):**
44
49
  ```bash
45
50
  docker run -d --name qdrant \
46
51
  -p 127.0.0.1:6333:6333 \
@@ -48,26 +53,39 @@ docker run -d --name qdrant \
48
53
  qdrant/qdrant
49
54
  ```
50
55
 
51
- **Embedding Provider** (choose one):
56
+ **Alternative:** Install natively from [qdrant.tech](https://qdrant.tech/documentation/guides/installation/)
57
+
58
+ ### 2. Embedding Provider
52
59
 
53
- - **Ollama** (recommended): Self-hosted embeddings
60
+ Choose one option:
54
61
 
62
+ **Option A: Ollama (recommended - free, runs locally)**
55
63
  ```bash
56
- # Install Ollama natively for GPU support (recommended)
57
- curl -fsSL https://ollama.ai/install.sh | sh
58
- # Visit https://ollama.ai for Windows installer
64
+ # Install Ollama
65
+ curl -fsSL https://ollama.ai/install.sh | sh # Linux/macOS
66
+ # For Windows: Download from https://ollama.ai
67
+
68
+ # Pull the embedding model
59
69
  ollama pull nomic-embed-text
70
+ ```
60
71
 
61
- # OR use Docker (CPU-only, slower)
62
- docker run -d --name ollama \
63
- -p 127.0.0.1:11434:11434 \
64
- -v ollama_data:/root/.ollama \
65
- ollama/ollama
72
+ **Option B: OpenAI (requires paid API key)**
73
+ ```bash
74
+ export OPENAI_API_KEY="your-api-key-here"
75
+ ```
76
+
77
+ ### Quick Verification
78
+
79
+ Test your setup:
80
+ ```bash
81
+ # Check Qdrant
82
+ curl http://localhost:6333/collections
66
83
 
67
- docker exec ollama ollama pull nomic-embed-text
84
+ # Check Ollama
85
+ curl http://localhost:11434/api/tags
68
86
  ```
69
87
 
70
- - **OpenAI**: Requires API key
88
+ If either fails, directory-indexer will show a helpful error with setup guidance.
71
89
 
72
90
  ## Configuration
73
91
 
@@ -78,8 +96,8 @@ Directory Indexer uses environment variables for configuration. Set these if you
78
96
  export QDRANT_ENDPOINT="http://localhost:6333"
79
97
  export OLLAMA_ENDPOINT="http://localhost:11434"
80
98
 
81
- # Optional database path (default: ~/.directory-indexer/data.db)
82
- export DIRECTORY_INDEXER_DB="/path/to/your/database.db"
99
+ # Optional data directory (default: ~/.directory-indexer)
100
+ export DIRECTORY_INDEXER_DATA_DIR="/path/to/data"
83
101
 
84
102
  # Optional Qdrant collection name (default: directory-indexer)
85
103
  # Note: Setting to "test" enables auto-cleanup for testing
@@ -101,7 +119,7 @@ export OLLAMA_API_KEY="your-ollama-key" # if using hosted Ollama
101
119
  "env": {
102
120
  "QDRANT_ENDPOINT": "http://localhost:6333",
103
121
  "OLLAMA_ENDPOINT": "http://localhost:11434",
104
- "DIRECTORY_INDEXER_DB": "/path/to/your/database.db"
122
+ "DIRECTORY_INDEXER_DATA_DIR": "/path/to/data"
105
123
  }
106
124
  }
107
125
  }
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "directory-indexer",
3
- "version": "0.0.5",
3
+ "version": "0.0.6",
4
4
  "description": "AI-powered directory indexing with semantic search for MCP servers",
5
5
  "main": "bin/directory-indexer.js",
6
6
  "bin": {