Solomd logo

Solomd

CommunityPopular
zhitongblog
solomd

Read, search, edit, and version-control any folder of Markdown notes via the SoloMD MCP server. 13 tools including AutoGit per-note history, semantic search, and write-with-sandbox via an accept/reject branch. Includes 11 starter agent recipes (weekly review, todo extract, link suggester, …).

Overview

Publisherzhitongblog
Repositorysolomd
Skill namesolomd
Stars
1.2K
Forks
80
Bundled files
12
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 12 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by zhitongblog on GitHub. Read the source before you install it.

Installation

Install the Solomd AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/zhitongblog/solomd/tree/main/marketplace/claude-code-skill/solomd
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/zhitongblog/solomd.git /tmp/solomd
mkdir -p .claude/skills
cp -r /tmp/solomd/marketplace/claude-code-skill/solomd .claude/skills/solomd
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Solomd in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Solomd on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Solomd is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

SoloMD — Markdown vault skill

When to use this skill

Trigger this skill when the user wants to:

  • Browse / read / search a folder of Markdown notes (their personal knowledge base, journal, project docs, etc.).
  • Maintain the vault — generate weekly summaries, extract TODOs across notes, suggest wikilinks, find orphan notes, etc.
  • Edit notes with version control — every write lands as an AutoGit commit (no network); past revisions are queryable via autogit_log / autogit_diff.
  • Share a note publicly via a solomd.app/share/?repo=…&path=… link that renders the file from a public GitHub repo (no SoloMD account required for the viewer).

If the user just wants generic file I/O on a directory, the built-in filesystem tools are simpler. This skill is specifically for Markdown-as-a-vault workflows.

Setup (one-time)

The skill assumes solomd-mcp is installed and registered as a user-scope MCP server (~/.claude.json). If /mcp doesn't list solomd, run:

bash
bash ~/.claude/skills/solomd/install.sh

This installs the binary (via cargo install solomd-mcp on macOS or downloading the prebuilt binary from https://github.com/zhitongblog/solomd/releases/latest on Linux / Windows) and registers a solomd server pointing at your notes folder via claude mcp add --scope user (falling back to patching ~/.claude.json directly when the claude CLI isn't on PATH).

Default mode is read-only. To enable the 3 write tools, re-register with --allow-write:

bash
claude mcp remove solomd -s user
claude mcp add --scope user solomd -- \
  ~/.claude/bin/solomd-mcp --workspace ~/your-notes --allow-write

Tools available

All exposed via the solomd: prefix.

Read tools (always available)

  • solomd:list_notes(workspace?, subdir?, limit=100) — list .md files.
  • solomd:read_note(workspace?, path) — full contents.
  • solomd:search(workspace?, query, mode="literal"|"regex", limit=200) — substring or regex search; returns matches with line numbers.
  • solomd:get_backlinks(workspace?, note_name) — every note that references [[<name>]].
  • solomd:list_tags(workspace?) — #tag aggregation with counts.
  • solomd:get_outline(workspace?, path) — H1/H2/H3 structure.
  • solomd:autogit_log(workspace?, path, limit=20) — per-note commit history. Each save in SoloMD is a commit.
  • solomd:autogit_diff(workspace?, path, from_revision, to_revision="HEAD") — diff between revisions of one note.
  • solomd:sync_status(workspace?) — git remote ahead/behind, dirty files, conflict state.
  • solomd:share_url(workspace?, path) — generate a public read-only solomd.app/share/… link (requires the workspace to be pushed to a public GitHub repo).
  • solomd:read_agent_trace(workspace?, run_id) — replay a past agent recipe run from its .solomd/agent-runs/<run-id>/trace.jsonl.

Write tools (need --allow-write on the server)

  • solomd:write_note(workspace?, path, content) — create or overwrite. Returns the diff against any prior version.
  • solomd:append_to_note(workspace?, path, content) — append; idempotent (won't duplicate a block already present at the bottom).
  • solomd:autogit_rollback(workspace?, path, revision) — restore a note to a past commit.

Multi-vault

If the server was started with multiple --workspace alias=/path flags, pass the workspace argument (e.g. workspace: "work" or workspace: "home") to disambiguate. Without one, the first registered workspace is the default.

Patterns

Browse before acting

When the user asks "do X across my notes", first call list_notes or search to confirm the scope. The vault might be 30 notes or 3,000 — your plan should match the size.

Always cite

When responding with information from notes, link them back as [[note-name]] (the wikilink format SoloMD uses). The user can click straight back to the source. Example:

Your three open architecture decisions are tracked in [[adr-001-storage]], [[adr-003-rate-limits]], and [[adr-007-streaming-api]].

Writes go through accept/reject — surface this

If the user is also running the SoloMD desktop app on the same vault, every write tool call lands on an agent/<recipe>/<run-id> AutoGit branch the user must accept in Settings → Recipes → Pending before it touches main. Tell the user this is happening so they know to check the app:

I've written the summary to weekly/2026-W22.md — it's pending on an AutoGit branch (agent/claude-code/abc123). Open SoloMD → Settings → Recipes → Pending to accept or reject before it lands on main.

If they're not running the app, the write lands on a branch they can inspect manually with git branch | grep agent/ and git merge / git branch -D as needed.

Watch the write cap

The server enforces a per-run write cap (default 5, hard ceiling 50). If you hit it, the call returns an error and no further writes succeed in that session. Plan multi-write actions in a single dispatch rather than fanning out one-at-a-time.

Starter recipes

This skill ships 11 reference YAML recipes at recipes/. They're the same files as the SoloMD Skill Pack zip — copy any of them into <workspace>/.solomd/agents/ to enable inside SoloMD (the app's recipe runner watches that directory).

For use from Claude Code directly, treat the YAML as documentation: adapt the prompt: block into the user's request and call the appropriate MCP tools yourself. Example — 01-weekly-review.yml becomes:

Use list_notes to find every daily/*.md from the past 7 days, read each, and write weekly/{ISO-week}.md with Themes / Decisions / Open threads sections, citing each daily as [[YYYY-MM-DD]].

Recipe catalog

FileTrigger (in-app)Pattern
01-weekly-review.ymlschedule (Sun 18:00)Summarise daily/ → weekly/YYYY-WW.md
02-todo-extract.ymlon-savePull - [ ] lines into todos.md
03-translate-zh-to-en.ymlon-tag-add #translateMirror translation alongside source
04-cjk-proofread.ymlon-saveLocal-Ollama punctuation + grammar
05-citation-cleanup.ymlon-tag-add #citeBibliography normalisation
06-meeting-notes-summary.ymlon-saveGenerate decisions / actions sections
07-link-suggester.ymlon-savePropose wikilinks via semantic search
08-daily-summary.ymlschedule (daily 22:00)Roll up today's edits
09-orphan-notes.ymlschedule (weekly)Find notes with no incoming wikilinks
10-on-commit-changelog.ymlon-commitMaintain CHANGELOG.md
11-tag-classifier.ymlon-saveSuggest tags from content

When NOT to use this skill

  • The user wants generic shell or filesystem ops on non-Markdown content — use the built-in Bash / Read / Edit tools.
  • The user is editing a single file they've already opened — direct Edit is more efficient than going through read_note / write_note.
  • The vault isn't actually a SoloMD vault and the user doesn't care about wikilinks / tags / AutoGit — the 5 SoloMD-only tools become no-ops (still useful for the other 8).

About SoloMD

SoloMD is a free, MIT-licensed Markdown editor. The MCP server you just installed is bundled with the desktop app — get the app for the visual Agent panel, the accept/reject UI, and AutoGit history navigation.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Solomd AI skill do?

Read, search, edit, and version-control any folder of Markdown notes via the SoloMD MCP server. 13 tools including AutoGit per-note history, semantic search, and write-with-sandbox via an accept/reject branch. Includes 11 starter agent recipes (weekly review, todo extract, link suggester, …).

Why use Solomd on TypingMind?

Because you install it once and use it with any model. Solomd is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Solomd in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/zhitongblog/solomd/tree/main/marketplace/claude-code-skill/solomd. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Solomd?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Solomd?

As many as you like. As long as a model supports skills, you can use Solomd with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Solomd AI skill free?

Yes. It is published on GitHub by zhitongblog under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇