qmd 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 skill's instructions expect a local 'qmd' CLI and automatic downloads of LLM/embedding models, but the package metadata does not declare those binaries, install steps, or model sources — that mismatch and the unspecified model downloads are concerning.

Purpose & Capability

The SKILL.md clearly expects the qmd CLI and local model runtimes (embeddinggemma-300M, qwen3-reranker-0.6b, Qwen3-0.6B). However, the registry metadata declares no required binaries, no install instructions, and no homepage/source. A user installing this skill would legitimately need qmd and model runtime support, so the omission is an incoherence that should be explained by the publisher.

Instruction Scope

The runtime instructions stay within the stated purpose (searching/indexing local files and returning snippets). They do imply the agent will read files in user-specified collections (expected for a search tool). The document also mentions 'models (auto-downloaded)' — the instructions do not specify where these downloads come from or whether they require network access or extra permissions.

Install Mechanism

There is no install spec (instruction-only), which is low-risk generally, but the SKILL.md claims models are auto-downloaded and run locally. The skill does not document the source of those model downloads (no URLs, releases, or registries). Unspecified automatic downloads of large model binaries increase risk (arbitrary network fetch & disk writes) and should be documented.

Credentials

The skill declares no required environment variables, credentials, or config paths and the instructions do not reference any secrets or unrelated env vars. That absence is proportionate to the stated purpose.

Persistence & Privilege

The skill is not marked 'always' and does not request persistent system privileges. As an instruction-only skill it does not modify other skills' configs or demand elevated privileges.

What to consider before installing

Before installing, verify these things with the publisher: (1) Confirm whether the qmd CLI is required and provide explicit 'required binaries' metadata (the SKILL.md assumes qmd is present). (2) Document install steps or a trusted source for the qmd binary and for the models (where the 'auto-downloaded' models come from—e.g., official releases, Hugging Face, or a vendor mirror). Unspecified model downloads can pull large binaries from arbitrary hosts and consume disk/network; ask how downloads are authenticated and where they are stored. (3) Understand which directories the skill will index and ensure you are comfortable granting the agent read access to those paths. (4) If you need higher assurance, request a homepage or source repo for the skill so you can inspect install scripts. If the publisher provides a clear install manifest and trusted model sources (or states qmd is preinstalled in your environment), this assessment could be downgraded to benign.

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.