agentleFS
ProjectAJ14ProjectAJ14/eklavya/skills/setup/SKILL.md

setup

Skill29 starsChanged 22 days agoMITView on GitHub

Check the Eklavya installation and database without changing your settings.

  • Installs packages

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

11 rules. 0 are common, 4 are distinctive, and 7 are very specific to this project.

  • 0

    Common

    Most of the 170 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 7

    Very specific

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

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

    Check that Eklavya works on this machine.

    line 9
  2. Distinctive:

    Do not ask the developer to choose settings or reset their saved choices.

    line 10
  3. Very specific:

    Run node --version. Node.js must be 22 or newer.

    line 16
  4. Very specific:

    Call get_config and report whether the server replied.

    line 20
  5. Very specific:

    If the call fails, check /plugin for errors.

    line 25
  6. Very specific:

    If get_config reports surface: "cowork", its successful reply confirms the server and database.

    line 32
  7. Distinctive:

    Do not tell the developer to install extra tools for this check.

    line 45
  8. Very specific:

    If questions are off, keep them off.

    line 51
  9. Distinctive:

    Do not grant permissions or change settings as part of this check.

    line 63
  10. Very specific:

    If the developer asks to enable commit enforcement, follow the commit gate guide.

    line 65
  11. Distinctive:

    Do not offer or install it during an ordinary setup check.

    line 68

“Common” is measured: kept by at least 6 in 10 of the 170 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

/eklavya:setup

Check that Eklavya works on this machine. Keep this to one short exchange. Do not ask the developer to choose settings or reset their saved choices. The installer supplies defaults. Change a setting only when the developer explicitly asks for that change.

Check Node and the server

Run node --version. Node.js must be 22 or newer. If it is missing or older, say how to install it on this platform and stop. The recommended npm install also needs npm, which the Node.js installer includes.

Call get_config and report whether the server replied. This call returns global_path and the effective configuration. Read that configuration before saying whether questions or memory are on; do not assume defaults describe an existing installation.

If the call fails, check /plugin for errors. From a terminal, the command npm install -g eklavya && eklavya install repairs or finishes the install. Then restart Claude Code. eklavya doctor can identify a broken runtime, database driver or plugin registration.

Check the database

If get_config reports surface: "cowork", its successful reply confirms the server and database. Do not check ~/.eklavya/knowledge.db from Cowork's shell: that shell runs in a separate VM and cannot see the user's local database.

Elsewhere, if sqlite3 is available, run:

sqlite3 ~/.eklavya/knowledge.db 'select domain, count(*) from concepts group by domain'

The starter domains are web-auth, react, node-backend and git. The installer creates and seeds this database. If sqlite3 is missing, use eklavya doctor instead; it is optional for ordinary questions and memory. Do not tell the developer to install extra tools for this check.

Say what happens next

With questions on, ask the developer to start a task. Point to /eklavya:progress for learning history and /eklavya:quiz for a question now. If questions are off, keep them off. When memory is on, mention /eklavya:memory for past work.

For a new install, memory summaries and prompt feedback use claude-haiku-5-5 through the developer's Claude Code login. Captured session evidence goes to that model; search and stored history stay local. Existing settings may use a different model or local summaries. If they ask about capture or privacy, link Memory.

The quiz side panel is on by default on supported Claude Code builds. If auto mode blocks it, link the ten auto-mode permissions. Do not grant permissions or change settings as part of this check.

If the developer asks to enable commit enforcement, follow the commit gate guide. The terminal Git hook also needs jq, sqlite3 and project-level enforcement. Do not offer or install it during an ordinary setup check.

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.

Reviews can't be read right now.

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.