Ea Connected Vault Setup logo

Ea Connected Vault Setup

Community
lnilya
ea-connected-vault-setup

Organize a sparse, mostly unconnected Markdown/Obsidian vault into a light topic hierarchy. Use when a user wants 5–10 topic root notes, a capitalized `Parents` list assigning notes to topic roots, an embedded Obsidian Base that lists each root's children, and meaningful body-level/`Related` wikilinks that make concept notes navigable without substantive rewriting.

Overview

Publisherlnilya
Repositoryeffortless-academic-skills
Skill nameea-connected-vault-setup
Stars
54
Forks
6
Bundled files
2
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.

  • 2 bundled files

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

  • Open source

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

Installation

Install the Ea Connected Vault Setup 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/lnilya/effortless-academic-skills/tree/main/skills/ea-connected-vault-setup
  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/lnilya/effortless-academic-skills.git /tmp/effortless-academic-skills
mkdir -p .claude/skills
cp -r /tmp/effortless-academic-skills/skills/ea-connected-vault-setup .claude/skills/ea-connected-vault-setup
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ea Connected Vault Setup 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 Ea Connected Vault Setup 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 Ea Connected Vault Setup 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.

EA Connected Vault Setup

Create a light navigation layer, not a new taxonomy or a content rewrite. Preserve every note's text, existing properties, filenames, folders, templates, query blocks, and links unless the requested linking requires a small additive edit.

Scope and safety

  • Work only on the user-designated corpus. Exclude configuration, templates, attachments, generated/cache folders, daily logs, and existing database/query files unless the user includes them.
  • Search filenames, aliases, frontmatter keys, links, and existing root/MOC notes before creating anything. Reuse a clearly matching root rather than duplicating it.
  • Use the exact, capitalized property names Parents and Related. Both are YAML/Markdown lists of wikilinks, even when they initially contain one item. Do not replace populated values; merge without changing their existing representation or casing.
  • In this initial setup, give each child exactly one Parents entry: its cluster root. Do not introduce multiple root assignments.
  • Do not infer a topical relationship from name similarity alone. Leave uncertain notes in a deliberately named holding root (for example, Unsorted concepts) and flag them.
  • This operation changes many files. First produce an inventory and proposed clustering; obtain confirmation before writing. After approval, apply only additive, minimal diffs and report every created/changed file.

1. Inventory and cluster proposal

  1. Read each in-scope Markdown note's title, aliases, frontmatter, headings, outbound links, and enough body text to establish its subject. Read linked context when it resolves ambiguity.
  2. Exclude the future root notes from the child count. Identify existing hubs, broad concepts, and duplicate/near-duplicate titles.
  3. Propose topic clusters that are semantically coherent and navigable:
    • Target 5–10 roots for a substantial collection.
    • Target roughly 8–15 child notes per root. Use fewer roots for a small corpus; split an oversized coherent cluster only when it has a meaningful internal division.
    • Give each root a specific, durable, plural-friendly name (for example, Plant functional traits, not Miscellaneous).
    • Assign every in-scope note exactly one proposed root, with a one-line reason for ambiguous assignments.
  4. Present the proposal as a compact table: root name, intended scope, child count, and any uncertain notes. Do not write yet.

2. Create roots and the child Base

After approval:

  1. Create the shared Base at __Support/Bases/Children by Parents.base unless the user specifies another support location. Copy the bundled [assets/Children by Parents.base](assets/Children by Parents.base) verbatim.

  2. Create a Roots/ folder at the vault root if it does not already exist. Create every new root note in Roots/, using that vault's root/topic template when one exists. If there is no template, use this minimal structure:

    md
    ---
    ---
    
    # Root name
    
    > [!note] AI-generated
    > Created by EA Connected Vault Setup as a navigation root. Review the cluster boundary before treating it as a settled taxonomy.
    
    ![[__Support/Bases/Children by Parents.base]]

    Keep the embed directly below the title/provenance. The Base resolves this to the root that embeds it, so the same file lists the correct children in every root. Its table displays the child Note first, followed by its Related list.

  3. If a reused root already exists outside Roots/, leave it in place; do not rename or move it. Add the Base embed only if it has no equivalent child listing, and do not overwrite or rearrange its content.

The Base contents are:

yaml
filters: 'Parents.contains(this.file)'
properties:
  file.name:
    displayName: Note
  Parents:
    displayName: Parents
  Related:
    displayName: Related
views:
  - type: table
    name: Children
    order:
      - file.name
      - Related

3. Assign each child

For every approved in-scope, non-root note:

  1. Add this list to the existing YAML frontmatter. Preserve ordering and formatting otherwise:

    yaml
    Parents:
      - "[[Root name]]"
  2. If the note lacks YAML frontmatter, add a minimal block containing this Parents list.

  3. Do not change an existing non-empty Parents list without flagging it in the review report. If it clearly points to an obsolete equivalent, ask before replacing it.

  4. Verify every Parents target resolves to a root note and every root Base has at least one expected child.

4. Make each note alive with discourse-advancing body links

Make a separate pass after parent assignment. A frontmatter list is only a navigation aid: it must not be the sole connection between notes. Review every substantive paragraph rather than aiming for a quota.

For each candidate target, read the target note and score the prospective link. Add it only when its overall case is strong; the purpose is to deepen the note's line of thought, not to make its graph denser.

Criterion0 points1 point2 points
1. Available targetNo suitable existing note.A broadly adjacent note exists.A specific existing note directly covers the concept, mechanism, method, evidence, or implication.
2. Contribution to discourseMerely interesting, decorative, or navigational.Helpful context.Advances the paragraph's reasoning: it defines a premise, supplies a mechanism, qualifies a claim, provides evidence, or opens a necessary next question.
3. Existing treatmentThe target is already linked nearby and another link adds no value.Mentioned/linked elsewhere in the note, but this paragraph needs a local entry point.Not already linked in the relevant context, or a deliberate repeated link materially improves navigation.
4. Minimal integrationRequires new claims or rewriting.Requires a short contextual line below the paragraph.Can be integrated by linking an existing phrase or making only a word-level edit.
5. Semantic precisionThe relationship is ambiguous or misleading.Plausible but indirect.The link label and target accurately express the exact relationship implied by the paragraph.
6. Reader utilityThe reader gains little by opening it.Useful background.Opening the target is likely to resolve a real conceptual, methodological, or evidential need created by this paragraph.
  • Add inline links for candidates scoring 9–12, and normally reject candidates scoring 0–6. Treat 7–8 as editorial judgment: retain only when the link clearly improves the reader's path through the argument.

  • Prefer direct inline links. Convert a natural, unlinked mention to [[Canonical note title|visible phrase]]. Make only the smallest word-level edit necessary for a grammatical, precise anchor; never alter a source claim or manufacture a sentence.

  • When a candidate clears the threshold but lacks a natural anchor, add a compact contextual line immediately below the relevant existing paragraph, for example:

    md
    *Related: [[Concept A]] and [[Concept B]].*

    This line must qualify or extend that paragraph without adding an unverified claim. Do not append detached link dumps or use these lines to satisfy a quota.

  • Add or merge the strongest body-level connections into a capitalized Related list in frontmatter. There is no fixed length: include only the compact set of links that remains useful as navigation. Related supplements body links rather than replacing them:

    yaml
    Related:
      - "[[Concept A]]"
      - "[[Concept B]]"
  • Preserve existing links and Related values. A second contextual link to the same note is valid only when it gives a later paragraph its own needed entry point. If no candidate clears the threshold, add none and list the note for human review.

5. Validate and report

  • Confirm that each child has exactly one resolving Parents entry.
  • Confirm that every Base file is valid YAML and its filter is exactly Parents.contains(this.file).
  • Check that every substantive paragraph was considered against the link score, that each added link meets the discourse/precision threshold, and that Related only supports rather than substitutes for those links.
  • Check for broken wikilinks introduced by this setup and for accidental links to the Base/template itself.
  • Report: Roots/ folder created/reused, roots created/reused, Base location, number of children assigned, number of inline/contextual links and Related lists added, uncertain notes, and anything intentionally untouched.
  • Clearly label the root notes as AI-created; do not add AI-generated prose to existing notes beyond the requested structural properties and links.

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 Ea Connected Vault Setup AI skill do?

Organize a sparse, mostly unconnected Markdown/Obsidian vault into a light topic hierarchy. Use when a user wants 5–10 topic root notes, a capitalized `Parents` list assigning notes to topic roots, an embedded Obsidian Base that lists each root's children, and meaningful body-level/`Related` wikilinks that make concept notes navigable without substantive rewriting.

Why use Ea Connected Vault Setup on TypingMind?

Because you install it once and use it with any model. Ea Connected Vault Setup 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 Ea Connected Vault Setup in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/lnilya/effortless-academic-skills/tree/main/skills/ea-connected-vault-setup. 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 Ea Connected Vault Setup?

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 Ea Connected Vault Setup?

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

Is the Ea Connected Vault Setup AI skill free?

Yes. It is published on GitHub by lnilya 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 👇