qmd-setup
Skill23 starsChanged 10 months agoMITView on GitHub
Set up @tobilu/qmd semantic search for mnemonic memories. Registers collections, builds indexes, and generates embeddings. Run this once per machine.
- Installs packages
In this repository9 other files it gives its agents.
- blackboardSkill
- coreSkill
- custodianSkill
- formatSkill
- integrateSkill
- ontologySkill
- qmd-reindexSkill
- searchSkill
- setupSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 22
Run the setup script:
- Very specific: line 29
Check prerequisites (Node.js >= 22, qmd installed)
- Very specific: line 35
Run qmd update (index) + qmd embed (embeddings)
- Distinctive: line 42
If you prefer to set up manually:
- Very specific: line 84
After adding new memories, re-index with /mnemonic:qmd-reindex or:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
QMD Setup Skill
Bootstrap @tobilu/qmd for semantic search over mnemonic memories.
Prerequisites
- Node.js >= 22 —
node --version - qmd CLI —
npm i -g @tobilu/qmd
Automated Setup
Run the setup script:
bash tools/qmd-setup.shThe script will:
- Check prerequisites (Node.js >= 22, qmd installed)
- Resolve
MNEMONIC_ROOTfrom~/.config/mnemonic/config.json - Discover memory roots dynamically:
{MNEMONIC_ROOT}/{org}/→ collectionmnemonic-{org}{MNEMONIC_ROOT}/default/→ collectionmnemonic-default.claude/mnemonic/→ collectionmnemonic-project
- Run
qmd update(index) +qmd embed(embeddings) - Validate with
qmd statusand a test search
Note: First
qmd embeddownloads ~2 GB of GGUF models.
Manual Setup
If you prefer to set up manually:
MNEMONIC_ROOT is resolved from ~/.config/mnemonic/config.json (defaults to
~/.local/share/mnemonic). Adjust the paths below to match your config:
# 1. Register collections (replace paths with your MNEMONIC_ROOT)
qmd collection add "${MNEMONIC_ROOT:-$HOME/.local/share/mnemonic}/zircote/" --name mnemonic-zircote
qmd collection add "${MNEMONIC_ROOT:-$HOME/.local/share/mnemonic}/default/" --name mnemonic-default
qmd collection add "$(git rev-parse --show-toplevel)/.claude/mnemonic/" --name mnemonic-project # if exists
# 2. Build index
qmd update
# 3. Generate embeddings (downloads models on first run)
qmd embed
# 4. Validate
qmd status
qmd search "test" -n 3After Setup
Use search commands:
| Command | Type | Requires |
qmd search "query" | BM25 keyword | qmd update |
qmd vsearch "query" | Semantic vector | qmd embed |
qmd query "query" | Hybrid (BM25 + vector) | Both |
Scope to a specific collection:
qmd search "auth" -c mnemonic-zircote # org memories only
qmd search "auth" -c mnemonic-project # this repo only
qmd search "auth" # all collectionsRe-indexing
After adding new memories, re-index with /mnemonic:qmd-reindex or:
qmd update && qmd embedIndexing is not automatic — run after captures or bulk imports.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

