agentleFS
spleenteospleenteo/maestro/.claude/skills/guide/SKILL.md

guide

Skill1 starsChanged 46 days agoMITView on GitHub

Answer the owner's questions about this orchestrator — how it works, what features exist, how to do something specific. Reads CLAUDE.md and the guides in howto/ to give grounded answers. Use when the owner types /guide, or says things like "I'm lost", "what can you do", "how do I", "how does this work", "remind me how to...", or similar confusion signals. Note: /help is a Claude Code built-in command and won't reach this skill — only /guide does.

21 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use grill, brainstorm and 6 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.
  • renatorozasSkill

    grill

    Adversarial code review. Don't let me ship until the changes pass your scrutiny.

    Also in 3 projects that use this

  • nicolasestremSkill

    brainstorm

    Structured idea generation and solution exploration with systematic analysis.

    Also in 3 projects that use this

  • BlockSkill

    plan

    Draft a TDD, PRD, or lightweight plan in the conversation without committing it.

    Also in 3 projects that use this

  • SentrySkill

    review

    Fixture review skill.

    Also in 4 projects that use this

  • Hugging FaceSkill

    implement

    Make tests pass. Invoke after /write-tests produces failing tests.

    Also in 3 projects that use this

  • NVIDIASkill

    research

    Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 16 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

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

    Use rg on frontmatter and section headings before reading bodies in full.

    line 37
  2. Very specific:

    When the answer isn't in the docs, say so plainly and offer two ways out: reason from the general pattern, or save the question as an idea …

    line 43
  3. Distinctive:

    Never invent a feature, a path or a behaviour.

    line 63
  4. Distinctive:

    Read-only, except the idea written on the owner's explicit consent.

    line 66

“Common” is measured: kept by at least 6 in 10 of the 16 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

Guide

The owner's way in when they're confused or exploring. It answers questions about the orchestrator itself, its features and conventions, by reading the project's own documentation. It never invents: when an answer isn't in the docs, it says so.

The skill is named guide because /help is a Claude Code built-in and never reaches a project skill.

Where the answers come from

Priority order:

  1. CLAUDE.md (repo root): the orchestrator's definition, first stop for any "how does the orchestrator handle X" question.
  2. howto/, one guide per topic (howto/README.md is the index):
    • 01-skills.md: adding, writing, retiring skills
    • 02-agents-and-hr.md: the HR agent, hiring, retiring, aliases
    • 03-customization.md: preferences, identity, file territories, reset
    • 04-memory-and-integrations.md: memory model, bin/mem, extending, integrations
    • 05-backup-and-sync.md: privacy, sync, symlinks, pointer skills, sub-apps
    • 06-configure-cal.md: the scheduler agent, data channels, routines
    • 07-warm-task-channel.md: an external task manager as the warm layer, the garbage collector
    • 08-markdown-discipline.md: frontmatter, tags, YAML safety, wikilinks
    • 09-semantic-memory.md: the optional semantic layer over the memory and the vault
    • 10-writing-register.md: the prose register, domains, tones, bin/register-check
    • 11-maestro-net.md: the channel between instances
    • 12-satellites.md: project repos attached to an instance
    • 13-archive.md: closing a vault folder into the archive
    • 14-agenda.md: the agenda pane, its two plugin scopes, language, migration from another marketplace
  3. private/preferences.md, only for questions about the owner's own setup ("what's my current language", "which territories do I have").
  4. The hub skills (.claude/skills/*/SKILL.md), when the question is about one skill's behaviour.

Use rg on frontmatter and section headings before reading bodies in full.

How to respond

  1. Detect the topic. A specific question ("how do I hire an agent?") goes straight to the relevant guide. A generic one (/guide, "I'm lost") gets the menu below, in the owner's default language, then waits for a choice.
  2. Answer short first: two to four sentences on what the owner asked, then the exact path (howto/01-skills.md, or CLAUDE.md → section name) for the full version, then one offer to go deeper.
  3. When the answer isn't in the docs, say so plainly and offer two ways out: reason from the general pattern, or save the question as an idea to document later. On "save as idea", write it with bin/mem idea "…" -t docs,guide,gap,<topic> and announce the write.

Menu for a generic question (translate into the owner's language):

What would you like help with?

  • How I work: my role, what I can and can't do on my own
  • Memory, logbook, TIL: where things get saved and how to read them back
  • Skills and agents: what they are, how to add one, how HR hires
  • Customization: my name, language, what I know about you
  • File territories and markdown: where I write and how files are tagged
  • Tasks: the memory db, an external task manager, Cal's questions
  • Writing register: how I write emails, documents and reports for you
  • Other instances and satellites: maestro-net, project repos attached to me
  • Backup, sync and sub-apps: what's private, what's in git, linking another project

Or ask me something more specific.

Rules

  • Never invent a feature, a path or a behaviour.
  • Short first, deep on demand: no full document dumped on the owner unless asked.
  • Cite the source: every substantive claim maps to a file path in the repo.
  • Read-only, except the idea written on the owner's explicit consent.
  • Respond in the owner's default language from private/preferences.md; when this folder isn't a configured instance, respond in English and point to /maestro:new-instance.
  • Stay in character: the orchestrator's voice from preferences.

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.