QMD - Quick Markdown Search

Local hybrid search for markdown notes and docs. Use when searching notes, finding related content, or retrieving documents from indexed collections.

Security Scan

The skill is internally consistent with a local Markdown search tool — it expects a qmd binary (installable via bun), indexes local files, and does not request credentials — but there are small metadata/instruction inconsistencies and it will download models/data to disk which you should be aware of.

Purpose & Capability

The SKILL.md describes a local markdown search tool that runs the qmd binary and may load local LLM models; that aligns with the stated purpose. However, the registry metadata for this skill lists no required binaries or env vars while the SKILL.md requires Bun and the qmd binary (and includes a bun install command). This mismatch is likely just an authoring gap but is worth noting.

Instruction Scope

Runtime instructions are narrowly scoped to installing/using qmd, adding collections, running search/embed/update commands, and optionally scheduling cron jobs. The skill instructs indexing local files and storing models/cache under ~/.cache/qmd/models (or XDG_CACHE_HOME). There are no instructions to read unrelated system config or exfiltrate secrets.

Install Mechanism

There is no install spec in the registry, but SKILL.md includes an install step using 'bun install -g https://github.com/tobi/qmd' (GitHub). Installing via Bun from a GitHub repo is common but should be treated as a network install from third-party code; model downloads at first run are also expected. This is moderate-risk compared with purely instruction-only skills but coherent for the tool's function.

Credentials

The skill does not request credentials or sensitive environment variables. It asks the user to ensure Bun is on PATH and notes where models/cache are stored. These are proportionate to a local search/indexing tool.

Persistence & Privilege

The skill is not always-enabled and is user-invocable. It suggests optional cron scheduling for index updates, but does not demand permanent elevated presence or modify other skills' configs. Autonomous invocation is allowed (platform default) but not combined with other high-risk indicators here.

Assessment

This skill appears to be a legitimate local markdown search helper that relies on the external 'qmd' binary and (optionally) local LLM models. Before installing, verify the upstream GitHub repo (https://github.com/tobi/qmd) so you know what code will be installed. Be aware that: (1) the first run can auto-download GGUF models into ~/.cache/qmd/models (large files and network activity), (2) installation uses 'bun install -g' which adds third-party code to your system, and (3) scheduling index/embedding jobs (cron or agent scheduler) will run qmd commands regularly against files you index. If you need stricter control, run qmd in a sandbox or VM, inspect the repository you install from, and confirm model download sources and network endpoints before allowing automatic downloads.

Local search engine for Markdown notes, docs, and knowledge bases. Index once, search fast.

When to use (trigger phrases)

• "search my notes / docs / knowledge base"
• "find related notes"
• "retrieve a markdown document from my collection"
• "search local markdown files"

Default behavior (important)

• Prefer qmd search (BM25). It's typically instant and should be the default.
• Use qmd vsearch only when keyword search fails and you need semantic similarity (can be very slow on a cold start).
• Avoid qmd query unless the user explicitly wants the highest quality hybrid results and can tolerate long runtimes/timeouts.

Prerequisites

• Bun >= 1.0.0
• macOS: brew install sqlite (SQLite extensions)
• Ensure PATH includes: $HOME/.bun/bin

Install Bun (macOS): brew install oven-sh/bun/bun

Install

bun install -g https://github.com/tobi/qmd

Setup

qmd collection add /path/to/notes --name notes --mask "**/*.md"
qmd context add qmd://notes "Description of this collection"  # optional
qmd embed  # one-time to enable vector + hybrid search

What it indexes

• Intended for Markdown collections (commonly **/*.md).
• In our testing, "messy" Markdown is fine: chunking is content-based (roughly a few hundred tokens per chunk), not strict heading/structure based.
• Not a replacement for code search; use code search tools for repositories/source trees.

Search modes

• qmd search (default): fast keyword match (BM25)
• qmd vsearch (last resort): semantic similarity (vector). Often slow due to local LLM work before the vector lookup.
• qmd query (generally skip): hybrid search + LLM reranking. Often slower than vsearch and may timeout.

Performance notes

• qmd search is typically instant.
• qmd vsearch can be ~1 minute on some machines because query expansion may load a local model (e.g., Qwen3-1.7B) into memory per run; the vector lookup itself is usually fast.
• qmd query adds LLM reranking on top of vsearch, so it can be even slower and less reliable for interactive use.
• If you need repeated semantic searches, consider keeping the process/model warm (e.g., a long-lived qmd/MCP server mode if available in your setup) rather than invoking a cold-start LLM each time.

Common commands

qmd search "query"             # default
qmd vsearch "query"
qmd query "query"
qmd search "query" -c notes     # Search specific collection
qmd search "query" -n 10        # More results
qmd search "query" --json       # JSON output
qmd search "query" --all --files --min-score 0.3

Useful options

• -n <num>: number of results
• -c, --collection <name>: restrict to a collection
• --all --min-score <num>: return all matches above a threshold
• --json / --files: agent-friendly output formats
• --full: return full document content

Retrieve

qmd get "path/to/file.md"       # Full document
qmd get "#docid"                # By ID from search results
qmd multi-get "journals/2025-05*.md"
qmd multi-get "doc1.md, doc2.md, #abc123" --json

Maintenance

qmd status                      # Index health
qmd update                      # Re-index changed files
qmd embed                       # Update embeddings

Keeping the index fresh

Automate indexing so results stay current as you add/edit notes.

• For keyword search (qmd search), qmd update is usually enough (fast).
• If you rely on semantic/hybrid search (vsearch/query), you may also want qmd embed, but it can be slow.

Example schedules (cron):

# Hourly incremental updates (keeps BM25 fresh):
0 * * * * export PATH="$HOME/.bun/bin:$PATH" && qmd update

# Optional: nightly embedding refresh (can be slow):
0 5 * * * export PATH="$HOME/.bun/bin:$PATH" && qmd embed

If your Clawdbot/agent environment supports a built-in scheduler, you can run the same commands there instead of system cron.

Models and cache

• Uses local GGUF models; first run auto-downloads them.
• Default cache: ~/.cache/qmd/models/ (override with XDG_CACHE_HOME).

Relationship to Clawdbot memory search

• qmd searches your local files (notes/docs) that you explicitly index into collections.
• Clawdbot's memory_search searches agent memory (saved facts/context from prior interactions).
• Use both: memory_search for "what did we decide/learn before?", qmd for "what's in my notes/docs on disk?".