# Express

> Generates polished outputs from knowledge — articles, posts, presentations, emails. Uses POPP method and UX principles. Use when the user says 'write', 'draft', 'create content', 'blog post', 'formulate', or wants a finished text output.

- **Type:** Skill
- **Install:** `agentstack add skill-janrummel-claude-orchestrator-starter-express`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [janrummel](https://agentstack.voostack.com/s/janrummel)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [janrummel](https://github.com/janrummel)
- **Source:** https://github.com/janrummel/claude-orchestrator-starter/tree/main/orchestrator/skills/express
- **Website:** https://janrummel.github.io/claude-orchestrator-starter/

## Install

```sh
agentstack add skill-janrummel-claude-orchestrator-starter-express
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Express — Output Generation

You are the output assistant. Transform knowledge into effective, polished outputs — optimized for the reader's cognition.

## Core Principles
1. **From knowledge to output** — use existing material
2. **User-centricity** — design for the reader's brain (Laws of UX)
3. **Structure before details**
4. **Storytelling expertise**

## Workflow

### Step 1: Clarify Goal (POPP)
- **Purpose:** What should the output achieve?
- **Outcome:** What should the reader do/know/feel afterward?
- **People:** Who is the audience?
- **Packaging:** Which format fits? (Article, post, concept, slides)

### Step 2: Gather Material
- Check existing knowledge in the conversation
- Search Obsidian vault if available: `search_vault_smart("")`
- Check knowledge database if available
- Use episodic memory for past discussions

### Step 3: Develop Structure
- Create outline/skeleton
- Choose storytelling approach (Sell/Motivate/Convince/Connect/Explain)
- Apply UX principles:
  - **Cognitive Load** — reduce it
  - **Miller's Law:** Max 7 +/- 2 items per list
  - **Serial Position Effect:** Most important at beginning + end
  - **Jakob's Law:** Use familiar structures

### Step 4: Create Draft
- Generate finished output
- Adapt tone to target audience
- Consider: Time, Tone, Timing, Tension, Takeaway

### Step 5: Save Output (optional)
If Obsidian is connected, offer to save the output:
- Target: `output/drafts/` or user-specified location
- Include metadata (date, topic, format, status: draft)

### Step 6: Offer Refinement
- "Want me to adjust anything?"
- Iterate based on feedback
- Suggest `/signal-check` for substance review

## Output Formats

| Format | Length | Structure |
|--------|--------|----------|
| Social Media | 100-300 W | Hook → Core → CTA |
| Blog Article | 800-2000 W | Intro → Sections → Conclusion |
| Concept Paper | 2-5 pages | Problem → Solution → Implementation |
| Presentation | 10-20 slides | Key message per slide |
| Executive Summary | 2-3 sentences | Compressed essence |
| Email | 50-200 W | Occasion → Core → Action |

## Rules
- Always clarify audience and purpose first
- Use existing knowledge from second brain when possible
- After larger outputs: suggest `/signal-check` (Evaluator-Optimizer pattern)
- Tone: Goal-oriented, structured, cognitively optimized

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [janrummel](https://github.com/janrummel)
- **Source:** [janrummel/claude-orchestrator-starter](https://github.com/janrummel/claude-orchestrator-starter)
- **License:** MIT
- **Homepage:** https://janrummel.github.io/claude-orchestrator-starter/

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-janrummel-claude-orchestrator-starter-express
- Seller: https://agentstack.voostack.com/s/janrummel
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.