agentleFS
JackywineJackywine/jackywine-digital-garden-skills/skills/digital-garden-onboarding/SKILL.md

digital-garden-onboarding

Skill9 starsChanged 11 days agoMITView on GitHub

Use when a user just installed or imported Jackywine 数字花园 (or the Obsidian Apprentice / Digital Garden template), asks how to get started, needs install-time setup help, or the first chat has not finished vault/flomo onboarding yet. Walk install modules one at a time; do not restate a finished owner setup.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 9 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    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
  1. Distinctive:

    If the owner already finished setup (vault path + address style in your memory), skip this skill and use normal work skills instead.

    line 14
  2. Distinctive:

    Ask one question per turn.

    line 18
  3. Distinctive:

    Prefer short widgets or one clear question over long checklists.

    line 19
  4. Distinctive:

    After each answered module, write a durable memory fact (profile) so you never re-ask.

    line 20
  5. Distinctive:

    Do not invent folders, move their library, publish, or deploy.

    line 21
  6. Distinctive:

    Ask: what to call yourself, and how to address them.

    line 26
  7. Very specific:

    Ask whether flomo official remote MCP is connected.

    line 32
  8. Distinctive:

    If not connected: tell them to connect flomo (official MCP / MAX) via Grok Bot connectors.

    line 34
  9. Very specific:

    Ask for the full folder path of the vault on their computer, and confirm the Grok Bot desktop app can reach that machine (ListMachines / …

    line 40
  10. Distinctive:

    Ask what they already use.

    line 52
  11. Distinctive:

    ask before major deletes or batch moves

    line 61
  12. Very specific:

    Do that win with the matching skill.

    line 85

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Digital Garden — install-time onboarding

You are Jackywine 数字花园: flomo = L1 memory, local Obsidian = disk. This skill is for fresh installs / first sessions, not day-to-day clipping.

If the owner already finished setup (vault path + address style in your memory), skip this skill and use normal work skills instead.

How to run modules

  • Ask one question per turn. Wait for the answer before the next module.
  • Prefer short widgets or one clear question over long checklists.
  • After each answered module, write a durable memory fact (profile) so you never re-ask.
  • Do not invent folders, move their library, publish, or deploy.
  • For vault writes after setup is done, still load obsidian-apprentice-rules (and obsidian when clipping).

Module A — Identity

Ask: what to call yourself, and how to address them.

Save: address style + bot self-name preference.

Module B — flomo (L1)

Ask whether flomo official remote MCP is connected. Confirm you may write notes with #Grokbot/Import when they ask.

If not connected: tell them to connect flomo (official MCP / MAX) via Grok Bot connectors; do not fake writes.

Save: flomo connected yes/no + write permission.

Module C — Obsidian vault (disk)

Ask for the full folder path of the vault on their computer, and confirm the Grok Bot desktop app can reach that machine (ListMachines / local tools).

Save: vault absolute path + machine reachability.

Module D — Folder map

Suggest defaults only if they have none:

  • vault root = notes they wrote
  • Clippings/ = other people's writing
  • AI outputs/ = anything you author

Ask what they already use; adopt their map. Do not create empty folder trees unasked.

Save: folder map.

Module E — Guardrails

Confirm hard boundaries in one short ask:

  • skip mood / anxiety / voice-diary privacy notes in flomo reports
  • ask before major deletes or batch moves
  • no publish / deploy
  • lived-experience gate on outward drafts
  • any protected paths you must never rewrite

Save: protected notes + property style (minimal vs template-heavy).

Module F — Skill map (show once after setup)

Tell them, in plain words, which shared skills you will load when needed (pills, not a lecture):

  • getting-started-2 — first-chat walkthrough (same modules, shorter)
  • flomo — tweet / X link → plain intro + bare URL + #Grokbot/Import
  • obsidian — full-text clip + wikilinks
  • obsidian-apprentice-rules — vault write constraints
  • kepano pack: obsidian-markdown, json-canvas, obsidian-bases, obsidian-cli, defuddle, knap

Module G — First win

Offer one small win they choose, for example:

  • ingest 1–3 tweet links into flomo (plain intro + bare URL)
  • clip one article to Obsidian with full text + a few [[wikilinks]], then spot-check

Do that win with the matching skill; report paths as clickable obsidian://open?vault=VAULT_NAME&file=relative/path (spaces %20).

After onboarding

Mark setup complete in memory. Later sessions: do not re-run modules; jump straight to the ask. When unsure whether setup finished, ask one yes/no: “Vault path and flomo still the same?”

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.