Skip to content
Back to skills

Mos Help

ASecurity

Explain marketing-os setup, architecture, routing, status, validation, and Claude Code or Codex wiring.

  • 4 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 23, 2026
ai-agents

Works with

  • claude code

Security analysis

A100/100

Scanned September 23, 2026

npx -y skills add the-vibe-marketing-lab/marketing-os --skill mos-help --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Mos Help?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Mos Help
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/the-vibe-marketing-lab-mos-help/badge)](https://www.skillsdirectory.com/skills/the-vibe-marketing-lab-mos-help)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: mos-help
description: Explain marketing-os setup, architecture, routing, status, validation, and Claude Code or Codex wiring.
---

# Help

Answer in plain business language using current facts rather than remembered command syntax.

## How to run this skill (interactive)

Help is a dialogue, not a lecture:

- If the question is broad or ambiguous, ask what the user is trying to do before explaining — a
  targeted answer beats a tour.
- After answering, offer the natural next step and ask whether they want to take it.

## Sources of truth

- Use `mos --help` and `mos <command> --help` for command syntax.
- For repository-specific questions, run `mos status . --json`.
- For health or discovery questions, run `mos doctor . --json`.
- Use `BRAIN.md` for routing and safety rules.

## System model

- `business/` is the sole source of business truth.
- `knowledge/` holds immutable sources and maintainable synthesized knowledge.
- `content/`, `campaigns/`, `reporting/`, and `outputs/` hold execution work.
- `mos` owns deterministic setup, facts, validation, and runtime wiring.
- The agent owns interviewing, judgment, synthesis, and writing.
- Claude Code and Codex consume generated copies of the same packaged skills.

## Common routes

- New or incomplete repository: `mos-onboard`.
- An existing folder that already holds a brain (legacy YAML config, or `BRAIN.md` beside
  `business/`): `mos attach <path> --plan`, then `--yes` — adopts it without rewriting content.
- Daily orientation: `mos-start`.
- Structural errors: `mos validate . --json`.
- Runtime discovery errors: preview and then apply `mos skills sync`.

Explain the outcome first. Include technical commands only when they help the user act or verify.
Do not claim that an uninstalled skill or unsupported command exists.

## Document contract

Every file you write under `business/`, `knowledge/`, `content/`, `campaigns/`, `reporting/`,
or `outputs/` opens with the frontmatter block defined in the repository's `CONTRACT.md`:
`title`, `type`, `description`, `date`, `status`, plus at least one of `sources`, `related`,
or `produced_by`. Deliverables must carry `sources:` — an output with no sources is not
finished. Emit the block as you write the file; never leave it for a later pass.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.