agentleFS
btrazzinibtrazzini/hippOS/skills/configure/SKILL.md

configure

Skill2 starsChanged 14 days agoMITView on GitHub

Sets up hippOS step by step, asking one question at a time - the hippOS folder in Google Drive, the person's profile, the first areas and, on a computer, the skills and the private copy of the repository. Use with /configure (or $configure in Codex), on first use, to add an area or to check the setup.

43 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use calendar, setup and 2 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.
  • jacobraskSkill

    calendar

    Interact with Apple Calendar via AppleScript. Use when the user asks to check calendar, view events, create events, manage schedule, find free time, or list calendars. Triggers include "my calendar", "my schedule", "calendar events", "create event", "add to calendar", "what's on my calendar", "free time", "available slots", "upcoming events", "today's events". Requires macOS with Calendar.app.

    Also in 3 projects that use this

  • VercelSkill

    setup

    Set up Vercel CLI and project configuration. Use when the user says "set up Vercel", "configure Vercel", "link to Vercel", or "vercel init".

    Also in 7 projects that use this

  • AnthropicSkill

    status

    Case status summary by audience — client-facing (plain language), internal (for the professor), or court-ready (formal caption format per local rules). Same facts, different framing and depth. Use when a student needs to update the client, brief the professor, or prepare a court status report.

    Also in 3 projects that use this

  • Microsoft AzureSkill

    build

    Build the TypeSpec Rust emitter from source. Run this after cloning the repo, after pulling changes, or when TypeScript source files have been modified.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 4 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 43 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Never fill in a field the person didn't give or confirm.

    line 10
  2. Very specific:

    Only in Claude Code or Codex, with a shell.

    line 11
  3. Distinctive:

    Check without changing anything, then tell the person in a few lines:

    line 14
  4. Distinctive:

    Ask, one at a time:

    line 20
  5. Distinctive:

    Ask which areas of life to start with (suggest common ones, such as work, personal, health, finances.

    line 30
  6. Very specific:

    Create: python3 memory.py area "<root>" <area> [alias ...], then fill AREA.md with the answers, in their words.

    line 37
  7. Distinctive:

    Add an area: step 3, then step 4.1.

    line 50
  8. Very specific:

    Check the setup: steps 4.2 to 4.6.

    line 52

“Common” is measured: kept by at least 6 in 10 of the 43 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

hippOS configure (v0.3.5)

Memory rules: protocol.md in this folder (read "Layers", "Root in Drive" and "Drive access"). Tools: memory.py in this folder (new and area build from the repository's templates/).

  • Talk in the language the person writes in. Ask one question at a time, with suggested answers when there are obvious ones (use the question tool if the environment has one). Before each step, say in one line what it does.
  • Never fill in a field the person didn't give or confirm. Never overwrite or delete a memory file: creating a memory or an area, and the profile's name, timezone and language, are the only writing outside sleep.
  • Only in Claude Code or Codex, with a shell. In chat: say that /configure runs in Claude Code (on the computer or at claude.ai/code) and stop. If SLEEP-IN-PROGRESS.md exists in the root, a sleep is running: say so and stop.

1. Where things stand

Check without changing anything, then tell the person in a few lines:

  • Drive: python3 memory.py root. A path: local mode, the root exists. Nothing: look for Google Drive for desktop ("My Drive" under ~/Library/CloudStorage/GoogleDrive-*/ on a Mac; a drive letter or ~/My Drive on Windows); with it, local mode, the root still to be created there (several Google accounts: ask which); without it, connector mode (search as in "Root": only a folder they own, directly in My Drive).
  • Memory: a root with core/PROFILE.md is an existing memory: go to step 5. Otherwise, steps 2 to 4; a root without it is an unfinished setup, and they complete what is missing.
  • Computer (local mode): are ~/.claude/skills/assistant, sleep and configure links to this repository? And git remote -v: an origin pointing to the public btrazzini/hippOS means there is no private copy yet.

2. Profile

Ask, one at a time:

  1. Language of replies and memory content (suggest the one they are writing in; e.g. en, pt-BR).
  2. What to call them (name).
  3. Timezone, IANA (suggest the computer's: readlink /etc/localtime on a Mac, the part after zoneinfo/; in the cloud the machine is usually UTC, so ask).
  4. Optional: a few lines about who they are that will hold for months (role, what they do, what matters to them).
  5. Optional: two or three preferences on how to be served (length of answers, tone, what to avoid).

Create: python3 memory.py new "<root>" "<name>" <timezone> <language>, where <root> is hippOS directly in My Drive. Then write the answers to 4 in the body of core/PROFILE.md (with [<name> date]) and to 5 in core/PREFERENCES.md, under the template's sections. Connector mode: build it in ~/.cache/hippos-configure/hippOS/ and, at the end of step 4, create the same folders and files in Drive: the hippOS folder directly in My Drive, folders before their files (mimeType: application/vnd.google-apps.folder), files as in "Drive access". Before creating each one, search for it by title and parent (the root: as in "Root"): if it exists, keep it and don't create another (the connector can't overwrite, and two files with the same title break the memory).

3. Areas

Ask which areas of life to start with (suggest common ones, such as work, personal, health, finances; one is enough). For each area, one question at a time:

  • Folder name: short, lowercase, hyphens (side-project); optional aliases.
  • Mission in one line, and what success looks like.
  • The assistant's role: what it does and what it doesn't.
  • Limits (what it must never do) and sources it should know (tools, repositories, documents).
  • GitHub lookups, only if they use GitHub for this area and gh auth status is logged in: which repositories, and whether to watch issues assigned to them, issues with a label, or PRs requesting their review (format in the ## Lookups comment of templates/area/AREA.md), plus their own and the people's logins and names.

Create: python3 memory.py area "<root>" <area> [alias ...], then fill AREA.md with the answers, in their words; a section they skipped keeps the template's ....

4. Finish

  1. python3 memory.py validate "<root>" with zero errors (fix what it flags), then python3 memory.py state "<root>" <area> --write for each area and python3 memory.py core "<root>" --write. Connector mode: run them on the local copy, upload everything, then run core again with --ids (a JSON of the created folders' ids by path, "hippOS/areas/<area>": "<id>", as sleep does) and upload MAP.md. Then delete ~/.cache/hippos-configure/: it holds a copy of the memory.
  2. Computer: skills not linked: offer to run ./install.sh from the repository.
  3. Private copy, needed for Claude Code in the cloud: if origin is the public hippOS, offer to create it; on a yes, with gh logged in, git remote rename origin upstream and gh repo create <name they choose> --private --source . --push. Never push to the public repository.
  4. Cloud and chat: the Google Drive connector turned on in Claude, with the Google account that holds hippOS; for chat, the zip from tools/package.sh.
  5. Nightly sleep, so each morning's opening starts consolidated: if the environment schedules tasks (a scheduled task in the Claude desktop app, or a cloud routine with /schedule), offer a /sleep every night at an hour they choose (suggest 03:00, their timezone). On a yes, create it with the prompt /sleep: on the computer, with the skills linked (4.2), and say the app runs it only while open (if closed, at the next launch); in the cloud, on the private copy (4.3), with the Drive connector (4.4). Elsewhere, they run /sleep themselves.
  6. Morning brief, so the assistant speaks first: where nightly sleep can be scheduled, offer /assistant with no area every morning at an hour they choose (suggest 07:30, after the sleep), created the same way. With an email connector, offer to also send it to their own address; that yes is the authorization.
  7. Close in up to 6 lines: what was created and the next step, /assistant <area>; conversations leave captures in the inbox, and /sleep consolidates them.

5. Existing memory

Say what is there (areas, profile fields) and ask what they want:

  • Add an area: step 3, then step 4.1.
  • Change name, timezone or language in core/PROFILE.md: only those fields, then python3 memory.py validate "<root>" core.
  • Check the setup: steps 4.2 to 4.6. Anything else in the memory (board, records, preferences) is a conversation with /assistant, which /sleep consolidates.

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.