qmd Local Search

Fast local search for markdown files, notes, and docs using qmd CLI. Use instead of `find` for file discovery. Combines BM25 full-text search, vector semantic search, and LLM reranking—all running locally. Use when searching for files, finding code, locating documentation, or discovering content in indexed collections.

Security Scan

The instructions match a local qmd CLI search tool, but the manifest omits the required qmd binary and references automatic large-model downloads from unspecified sources — these mismatches and unspecified network activity warrant caution.

Purpose & Capability

The SKILL.md clearly expects a local 'qmd' CLI to be present and describes collection/index operations. However, the skill metadata claims no required binaries or install steps. That mismatch (instructions assuming a binary that the manifest doesn't require) is incoherent and should be clarified.

Instruction Scope

Instructions are limited to running qmd to index and search local files, which fits the stated purpose. However wording like 'pull relevant snippets before answering questions' is broad and could lead to indexing or exposing sensitive local content unless the user scopes collections carefully.

Install Mechanism

There is no install spec, but SKILL.md states models are 'auto-downloaded' (embeddinggemma-300M, qwen3-reranker-0.6b, Qwen3-0.6B). The source, host, and trustworthiness of those downloads are unspecified — implicit large model downloads from unknown endpoints are a risk.

Credentials

The skill declares no required environment variables, credentials, or config paths, which is proportionate to a local-search tool. No environment secrets are requested.

Persistence & Privilege

The skill is instruction-only and not marked 'always', but manifest flags for model invocation are unset (default allows the model to invoke the skill). That means the agent could call the qmd commands autonomously if qmd exists — consider whether autonomous invocation is acceptable.

What to consider before installing

This skill appears to document how to use a local 'qmd' CLI, but there are two issues you should address before installing/using it: (1) The manifest does not list 'qmd' as a required binary even though all instructions assume it exists — confirm you will install qmd from a trusted source (official repo or package) and that the agent runtime will have the binary on PATH. (2) The SKILL.md says several models will be 'auto-downloaded' but doesn't state where from — verify the model hosting, bandwidth/storage requirements, and trustworthiness (prefer official model repos or allow only offline/air-gapped model use). Also: restrict which folders are added as collections to avoid indexing sensitive data, and consider disabling autonomous model invocation or requiring explicit user confirmation before the agent runs qmd commands. If the skill author can supply a clear install spec and model download sources, the incoherence would be largely resolved.

qmd — Fast Local Markdown Search

When to Use

• Finding files — use instead of find across large directories (avoids hangs)
• Searching notes/docs — semantic or keyword search in indexed collections
• Code discovery — find implementations, configs, or patterns
• Context gathering — pull relevant snippets before answering questions

Quick Reference

Search (most common)

# Keyword search (BM25)
qmd search "alpaca API" -c projects

# Semantic search (understands meaning)
qmd vsearch "how to implement stop loss"

# Combined search with reranking (best quality)
qmd query "trading rules for breakouts"

# File paths only (fast discovery)
qmd search "config" --files -c kell

# Full document content
qmd search "pattern detection" --full --line-numbers

Collections

# List collections
qmd collection list

# Add new collection
qmd collection add /path/to/folder --name myproject --mask "*.md,*.py"

# Re-index after changes
qmd update

Get Files

# Get full file
qmd get myproject/README.md

# Get specific lines
qmd get myproject/config.py:50 -l 30

# Get multiple files by glob
qmd multi-get "*.yaml" -l 50 --max-bytes 10240

Output Formats

• --files — paths + scores (for file discovery)
• --json — structured with snippets
• --md — markdown formatted
• -n 10 — limit results

Tips

1. Always use collections (-c name) to scope searches
2. Run qmd update after adding new files
3. Use qmd embed to enable vector search (one-time, takes a few minutes)
4. Prefer qmd search --files over find for large directories

Models (auto-downloaded)

• Embedding: embeddinggemma-300M
• Reranking: qwen3-reranker-0.6b
• Generation: Qwen3-0.6B

All run locally — no API keys needed.