agentleFS
slogsdonslogsdon/loop-and-gate-foundation/skills/setup/SKILL.md

setup

Skill0 starsChanged 4 months agoMITView on GitHub

Place and scaffold the memory vault on first run. Use once right after installing — as a plugin (ask "set up my vault", /setup, or /skill:setup on pi) or a fresh clone — or whenever the SessionStart hook reports no vault is configured. Prompts for the location, then records the path so memory loads every session after.

191 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use add-voice-transcription, add-discord and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • juancruzmunozalbeloSkill

    add-voice-transcription

    Add voice message transcription to NanoClaw using OpenAI's Whisper API. Automatically transcribes WhatsApp voice notes so the agent can read and respond to them.

    Also in 5 projects that use this

  • MicrosoftSkill

    add-discord

    Add Discord bot channel integration via Chat SDK.

    Also in 11 projects that use this

  • MicrosoftSkill

    add-slack

    Add Slack channel integration via Chat SDK.

    Also in 11 projects that use this

  • juancruzmunozalbeloSkill

    add-telegram-swarm

    Add Agent Swarm (Teams) support to Telegram. Each subagent gets its own bot identity in the group. Requires Telegram channel to be set up first (use /add-telegram). Triggers on "agent swarm", "agent teams telegram", "telegram swarm", "bot pool".

    Also in 7 projects that use this

  • +8 more from microsoft

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 3 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 171 edited versions we read keep these.

  • 3

    Distinctive

    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. Very specific:

    If neither resolves, tell the user the plugin isn't installed where expected and stop.

    line 34
  2. Distinctive:

    Run the script with the chosen path as its one argument:

    line 50
  3. Very specific:

    Never reimplement the scaffold logic here.

    line 64
  4. Distinctive:

    Never pick the location silently.

    line 66
  5. Distinctive:

    Run it once. It's a first-run step, not a session ritual.

    line 69
  6. Very specific:

    Keep setup.sh's git gate: the script *prints* the vault git-init command rather than running it.

    line 70

“Common” is measured: kept by at least 6 in 10 of the 171 edited versions of this file we could read. “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

Setup

One job: give the agent a memory vault to read and write. Ask where it should live, scaffold the folder structure there, and record the path so the SessionStart hook and every skill resolve the same vault afterward. Run once per machine. Idempotent — it never overwrites files you've already started.

This exists mainly for the plugin install path, where there is no scripts/setup.sh in your working directory to run — the script ships inside the read-only plugin cache. This skill locates that bundled script, asks you where the vault goes, and runs it. On a clone it does the same with the local script.

Steps

  1. Locate the bundled setup script. On pi the Foundation extension exports LOOP_GATE_ROOT; otherwise prefer the working copy, then fall back to the Claude Code plugin cache (the version-pinned path a plugin user can't easily type):

    script="${LOOP_GATE_ROOT:-.}/scripts/setup.sh"
    if [ ! -x "$script" ]; then
      script=$(find "${CLAUDE_CONFIG_DIR:-$HOME/.claude}/plugins" \
        -path '*loop-and-gate-foundation*/scripts/setup.sh' 2>/dev/null | sort -V | tail -1)
    fi

    If neither resolves, tell the user the plugin isn't installed where expected and stop. Don't scaffold by hand — setup.sh is the one source of truth for what a vault contains.

  2. Ask where the vault should live — this is a gate, don't decide it for the user. Check the OS first (uname). Ask with AskUserQuestion (Claude Code) or ask_user_question (pi), as a location choice:
    • On macOS, put the Obsidian iCloud folder first and mark it (Recommended) — ~/Library/Mobile Documents/iCloud~md~obsidian/Documents/SecondBrain syncs the vault to Obsidian on iPhone/iPad for free. Silently adopting a location is what broke earlier setups.
    • Offer a plain local folder (~/second-brain, no cloud sync).
    • Offer "an existing Obsidian vault" — the user pastes an absolute path and the scaffold lands inside it. Off macOS, drop the iCloud option and recommend the local folder.
  3. Run the script with the chosen path as its one argument:

    "$script" "<chosen-vault-path>"
  4. Report where the vault landed (the script prints and records the path) and the next steps: start a session, run profile-interview to teach it your voice, or run add-kits to add the rest of the Loop & Gate stack (the Build, Grow, and Accountability kits). Offer add-kits here but don't run it inline — it's a separate gated skill, and vault setup is done.

Rules

  • Never reimplement the scaffold logic here. This skill's only job is to ask the location and run setup.sh. Two copies of the folder layout would drift.
  • Never pick the location silently. The location is the user's call — that is the whole reason this skill asks the user instead of trusting the script's auto-detection.
  • Run it once. It's a first-run step, not a session ritual.
  • Keep setup.sh's git gate: the script prints the vault git-init command rather than running it. Don't git-init the vault for the user unasked.

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.