Skip to main content
Display comprehensive status of the QMD index, collections, embeddings, and system resources.

​
Usage

qmd status
No options or arguments.

​
Output Sections

​
Index Information

QMD Status

Index: /home/user/.cache/qmd/index.sqlite
Size:  42.3 MB
MCP:   running (PID 12345)
Shows:
  • Index file path
  • Database size on disk
  • MCP daemon status (if running)

​
Documents

Documents
  Total:    1,247 files indexed
  Vectors:  1,189 embedded
  Pending:  58 need embedding (run 'qmd embed')
  Updated:  2h ago
Shows:
  • Total active documents
  • Documents with embeddings
  • Documents needing embeddings (yellow warning if >0)
  • Time since last update

​
Collections

Collections
  notes (qmd://notes/)
    Pattern:  **/*.md
    Files:    423 (updated 2h ago)
    Contexts: 3
      /:           Personal notes and documentation
      /meetings:   Weekly team meeting notes
      /projects:   Active project documentation
  
  journals (qmd://journals/)
    Pattern:  **/*.md
    Files:    824 (updated 5d ago)
    Contexts: 1
      /:           Daily journal entries
For each collection:
  • Name and virtual path
  • Glob pattern
  • File count and last update time
  • Context count and preview (first 60 chars)

​
Examples

Examples
  # List files in a collection
  qmd ls notes
  # Get a document
  qmd get qmd://notes/readme.md
  # Search within a collection
  qmd search "query" -c notes

​
Models

Models
  Embedding:   https://huggingface.co/google/gemma-2-2b-it
  Reranking:   https://huggingface.co/Qwen/Qwen2.5-3B-Instruct
  Generation:  https://huggingface.co/Qwen/Qwen2.5-3B-Instruct
Links to Hugging Face model pages.

​
Device

Device
  GPU:      NVIDIA GeForce RTX 3080 (offloading: yes)
  Devices:  NVIDIA GeForce RTX 3080
  VRAM:     8.2 GB free / 10.0 GB total
  CPU:      8 math cores
Or if no GPU: 
Device
  GPU:      none (running on CPU — models will be slow)
  Tip: Install CUDA, Vulkan, or Metal support for GPU acceleration.
  CPU:      8 math cores

​
Tips

Tips
  Add context to collections for better search results: notes, journals
    qmd context add qmd://<name>/ "What this collection contains"
    qmd context add qmd://<name>/meeting-notes "Weekly team meeting notes"
  Add update commands to keep collections fresh: notes, journals
    qmd collection update-cmd <name> 'git stash && git pull --rebase --ff-only && git stash pop'
Context-aware tips shown when:
  • Collections missing context
  • Collections missing update commands (if more than 1 collection)

​
Health Warnings

Status automatically checks index health: Many documents need embedding (≥10%): 
Warning: 156 documents (42%) need embeddings. Run 'qmd embed' for better results.
Few documents need embedding (<10%): 
Tip: 12 documents need embeddings. Run 'qmd embed' to index them.
Stale index (>= 14 days): 
Tip: Index last updated 21 days ago. Run 'qmd update' to refresh.

​
MCP Daemon Detection

Checks ~/.cache/qmd/mcp.pid for running daemon:
  • Shows PID if running
  • Silent if not running
  • Auto-cleans stale PID files

​
Use Cases

Quick health check: 
qmd status
Before searching: 
qmd status  # Check if embeddings are current
qmd embed   # If needed
qmd query "search terms"
After adding collection: 
qmd collection add ~/notes
qmd status  # See new collection
qmd embed   # Generate embeddings
Troubleshooting: 
qmd status  # Check GPU, model paths, collection counts

qmd collection list

List collections only

qmd embed

Generate embeddings

qmd update

Refresh collections

qmd context list

List all contexts

Build docs developers (and LLMs) love

Get started for freeTalk to us