# Prompt Build

> >

- **Type:** Skill
- **Install:** `agentstack add skill-agricidaniel-claude-prompts-prompt-build`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [AgriciDaniel](https://agentstack.voostack.com/s/agricidaniel)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [AgriciDaniel](https://github.com/AgriciDaniel)
- **Source:** https://github.com/AgriciDaniel/claude-prompts/tree/main/skills/prompt-build
- **Website:** https://www.skool.com/ai-marketing-hub-pro/

## Install

```sh
agentstack add skill-agricidaniel-claude-prompts-prompt-build
```

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

## About

# Prompt Builder

Build production-quality AI prompts using patterns extracted from 2,500+ curated prompts.

## Build Workflow

### Step 1: Determine Intent

Ask the user (if not already clear):
1. **Output type**: Image, Video, or Text?
2. **Target model**: Which AI model? (Midjourney, Flux, DALL-E, Leonardo, Sora, etc.)
3. **Subject**: What should the prompt create?
4. **Style/mood**: Any specific aesthetic?

### Step 2: Find Reference Prompts

Search the database for similar prompts to use as inspiration:
```bash
python3 {PROMPT_ENGINE_DIR}/scripts/search_prompts.py "SUBJECT" --model MODEL --limit 5
```

### Step 3: Construct the Prompt

Use this universal structure (adapt per model):

**Image Prompts:**
```
[Subject/Action] + [Setting/Environment] + [Lighting/Mood] + [Camera/Lens] +
[Style Reference] + [Technical Parameters]
```

**Video Prompts:**
```
[Camera Movement] + [Subject/Action] + [Setting] + [Lighting] +
[Duration/Speed] + [Style] + [Technical]
```

**Text Prompts:**
```
[Role/Context] + [Task] + [Constraints] + [Format] + [Tone] + [Examples]
```

### Step 4: Apply Model-Specific Syntax

Load `{PROMPT_ENGINE_DIR}/references/model-guide.md` for model-specific formatting rules:
- **Midjourney**: Use `--ar`, `--v`, `--style`, `--s`, `--chaos`
- **Flux**: Natural language, detailed descriptions work best
- **DALL-E**: Direct, descriptive, avoid negative prompts
- **Leonardo AI**: Use model-specific keywords (Phoenix, Alchemy)
- **Sora**: Camera movements, scene transitions, temporal descriptions

### Step 5: Review and Refine

Present the prompt with:
1. The full prompt text
2. Recommended parameters/settings
3. Similar prompts from the database for comparison
4. Suggested variations (3 alternatives)

## Quality Gates

- Prompt must be > 30 characters
- Must include subject + at least one modifier (style, lighting, or camera)
- Must use model-appropriate syntax
- No conflicting instructions (e.g., "realistic" + "cartoon")

## Source & license

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

- **Author:** [AgriciDaniel](https://github.com/AgriciDaniel)
- **Source:** [AgriciDaniel/claude-prompts](https://github.com/AgriciDaniel/claude-prompts)
- **License:** MIT
- **Homepage:** https://www.skool.com/ai-marketing-hub-pro/

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-agricidaniel-claude-prompts-prompt-build
- Seller: https://agentstack.voostack.com/s/agricidaniel
- 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%.