How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

frankx-ai-prompts - Prompt

Name: frankx-ai-prompts
Author: frankxai

Prompt file imported from frankxai/frankx.ai-vercel-website (.claude/commands/frankx-ai-prompts.md). Copyright stays with the author.

FrankX.AI Prompt Library Management

Location: /lib/prompts.ts Live at: https://frankx.ai/prompt-library

Current Library Stats

The prompt library contains 25+ prompts across 12 categories serving creators, executives, and developers.

Adding a New Prompt

Step 1: Determine Category

Category ID	Name	Focus
`task-specific`	Task-Specific	Operational playbooks
`ai-optimization`	AI Optimization	Model tuning, multi-agent
`business-applications`	Business Applications	Strategy, go-to-market
`creative-workflows`	Creative Workflows	Narrative, concepts
`technical-development`	Technical Development	Engineering, QA
`educational-content`	Educational Content	Curriculum, cohorts
`music-audio`	Music & Audio	Suno, DAW, sonic
`visual-design`	Visual Design	Midjourney, DALL-E
`data-analysis`	Data Analysis	Dashboards, research
`communication`	Communication	Storytelling
`productivity`	Productivity	Rituals, leverage
`interactive-experiences`	Interactive Experiences	Games, experiences

Step 2: Prompt Structure

Add to the prompts array in /lib/prompts.ts:

{
  id: 'unique-slug-id',
  title: 'Prompt Title',
  description: 'One sentence description of what this prompt does',
  category: 'category-id',
  tool: 'Claude' | 'ChatGPT' | 'Midjourney' | 'Suno' | 'DALL-E' | 'Notion' | 'Other',
  difficulty: 'Beginner' | 'Intermediate' | 'Advanced',
  usageCount: 0,
  rating: 4.5,
  prompt: `
The full prompt text here.
Can be multiline.
Use markdown formatting.
`,
  expectedOutput: 'What the user should expect to receive',
  tips: [
    'Tip 1 for best results',
    'Tip 2 for customization',
  ],
  tags: ['tag1', 'tag2', 'tag3'],
}

Step 3: Quality Checklist

Before adding:

Tested and produces good results
Clear, actionable instructions
Includes expected output description
2-3 useful tips
Proper difficulty rating
Relevant tags for discoverability

Step 4: Category Balance

Check category distribution:

# Count prompts per category
grep -c "category:" lib/prompts.ts

Priority gaps to fill:

More Suno/music prompts (Frequency Alchemist specialty)
Creator-focused AI workflow prompts
Enterprise governance prompts

Prompt Ideas from FrankX Content

Extract prompts from:

Blog articles (frameworks become prompts)
Product offerings (toolkit prompts)
Daily workflows (operational prompts)

From Intelligence Atlas:

Multi-agent orchestration prompts
Governance assessment prompts
Creative studio setup prompts

From Vibe OS:

Session intention prompts
Frequency mapping prompts
Sonic identity prompts

Updating Existing Prompts

Find prompt by ID in /lib/prompts.ts
Update content, keeping ID stable
Increment version if major change
Update usageCount based on analytics

Bulk Operations

// Update all ratings
prompts.forEach(p => p.rating = calculateNewRating(p.id));

// Add new category
const newCategory = { id: 'new-cat', name: 'New Category', icon: 'Icon' };

Search Index Update

After adding prompts:

npm run gen:search

Ready to add or update prompts. What prompt are we working on?

Install frankx-ai-prompts and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

# FrankX.AI Prompt Library Management **Location**: `/lib/prompts.ts` **Live at**: https://frankx.ai/prompt-library ## Current Library Stats The prompt library contains 25+ prompts across 12 categories serving creators, executives, and developers. ## Adding a New Prompt ### Step 1: Determine Category | Category ID | Name | Focus | |-------------|------|-------| | `task-specific` | Task-Specific | Operational playbooks | | `ai-optimization` | AI Optimization | Model tuning, multi-agent | | `business-applications` | Business Applications | Strategy, go-to-market | | `creative-workflows` | Creative Workflows | Narrative, concepts | | `technical-development` | Technical Development | Engineering, QA | | `educational-content` | Educational Content | Curriculum, cohorts | | `music-audio` | Music & Audio | Suno, DAW, sonic | | `visual-design` | Visual Design | Midjourney, DALL-E | | `data-analysis` | Data Analysis | Dashboards, research | | `communication` | Communication | Storytelling | | `productivity` | Productivity | Rituals, leverage | | `interactive-experiences` | Interactive Experiences | Games, experiences | ### Step 2: Prompt Structure Add to the prompts array in `/lib/prompts.ts`: ```typescript { id: 'unique-slug-id', title: 'Prompt Title', description: 'One sentence description of what this prompt does', category: 'category-id', tool: 'Claude' | 'ChatGPT' | 'Midjourney' | 'Suno' | 'DALL-E' | 'Notion' | 'Other', difficulty: 'Beginner' | 'Intermediate' | 'Advanced', usageCount: 0, rating: 4.5, prompt: ` The full prompt text here. Can be multiline. Use markdown formatting. `, expectedOutput: 'What the user should expect to receive', tips: [ 'Tip 1 for best results', 'Tip 2 for customization', ], tags: ['tag1', 'tag2', 'tag3'], } ``` ### Step 3: Quality Checklist Before adding: - [ ] Tested and produces good results - [ ] Clear, actionable instructions - [ ] Includes expected output description - [ ] 2-3 useful tips - [ ] Proper difficulty rating - [ ] Relevant tags for discoverability ### Step 4: Category Balance Check category distribution: ```bash # Count prompts per category grep -c "category:" lib/prompts.ts ``` Priority gaps to fill: - More Suno/music prompts (Frequency Alchemist specialty) - Creator-focused AI workflow prompts - Enterprise governance prompts ## Prompt Ideas from FrankX Content Extract prompts from: - Blog articles (frameworks become prompts) - Product offerings (toolkit prompts) - Daily workflows (operational prompts) ### From Intelligence Atlas: - Multi-agent orchestration prompts - Governance assessment prompts - Creative studio setup prompts ### From Vibe OS: - Session intention prompts - Frequency mapping prompts - Sonic identity prompts ## Updating Existing Prompts 1. Find prompt by ID in `/lib/prompts.ts` 2. Update content, keeping ID stable 3. Increment version if major change 4. Update `usageCount` based on analytics ## Bulk Operations ```typescript // Update all ratings prompts.forEach(p => p.rating = calculateNewRating(p.id)); // Add new category const newCategory = { id: 'new-cat', name: 'New Category', icon: 'Icon' }; ``` ## Search Index Update After adding prompts: ```bash npm run gen:search ``` **Ready to add or update prompts. What prompt are we working on?**

{ "prompt": "# FrankX.AI Prompt Library Management\n\n**Location**: `/lib/prompts.ts`\n**Live at**: https://frankx.ai/prompt-library\n\n## Current Library Stats\n\nThe prompt library contains 25+ prompts across 12 categories serving creators, executives, and developers.\n\n## Adding a New Prompt\n\n### Step 1: Determine Category\n\n| Category ID | Name | Focus |\n|-------------|------|-------|\n| `task-specific` | Task-Specific | Operational playbooks |\n| `ai-optimization` | AI Optimization | Model tuning, multi-agent |\n| `business-applications` | Business Applications | Strategy, go-to-market |\n| `creative-workflows` | Creative Workflows | Narrative, concepts |\n| `technical-development` | Technical Development | Engineering, QA |\n| `educational-content` | Educational Content | Curriculum, cohorts |\n| `music-audio` | Music & Audio | Suno, DAW, sonic |\n| `visual-design` | Visual Design | Midjourney, DALL-E |\n| `data-analysis` | Data Analysis | Dashboards, research |\n| `communication` | Communication | Storytelling |\n| `productivity` | Productivity | Rituals, leverage |\n| `interactive-experiences` | Interactive Experiences | Games, experiences |\n\n### Step 2: Prompt Structure\n\nAdd to the prompts array in `/lib/prompts.ts`:\n\n```typescript\n{\n id: 'unique-slug-id',\n title: 'Prompt Title',\n description: 'One sentence description of what this prompt does',\n category: 'category-id',\n tool: 'Claude' | 'ChatGPT' | 'Midjourney' | 'Suno' | 'DALL-E' | 'Notion' | 'Other',\n difficulty: 'Beginner' | 'Intermediate' | 'Advanced',\n usageCount: 0,\n rating: 4.5,\n prompt: `\nThe full prompt text here.\nCan be multiline.\nUse markdown formatting.\n`,\n expectedOutput: 'What the user should expect to receive',\n tips: [\n 'Tip 1 for best results',\n 'Tip 2 for customization',\n ],\n tags: ['tag1', 'tag2', 'tag3'],\n}\n```\n\n### Step 3: Quality Checklist\n\nBefore adding:\n- [ ] Tested and produces good results\n- [ ] Clear, actionable instructions\n- [ ] Includes expected output description\n- [ ] 2-3 useful tips\n- [ ] Proper difficulty rating\n- [ ] Relevant tags for discoverability\n\n### Step 4: Category Balance\n\nCheck category distribution:\n```bash\n# Count prompts per category\ngrep -c \"category:\" lib/prompts.ts\n```\n\nPriority gaps to fill:\n- More Suno/music prompts (Frequency Alchemist specialty)\n- Creator-focused AI workflow prompts\n- Enterprise governance prompts\n\n## Prompt Ideas from FrankX Content\n\nExtract prompts from:\n- Blog articles (frameworks become prompts)\n- Product offerings (toolkit prompts)\n- Daily workflows (operational prompts)\n\n### From Intelligence Atlas:\n- Multi-agent orchestration prompts\n- Governance assessment prompts\n- Creative studio setup prompts\n\n### From Vibe OS:\n- Session intention prompts\n- Frequency mapping prompts\n- Sonic identity prompts\n\n## Updating Existing Prompts\n\n1. Find prompt by ID in `/lib/prompts.ts`\n2. Update content, keeping ID stable\n3. Increment version if major change\n4. Update `usageCount` based on analytics\n\n## Bulk Operations\n\n```typescript\n// Update all ratings\nprompts.forEach(p => p.rating = calculateNewRating(p.id));\n\n// Add new category\nconst newCategory = { id: 'new-cat', name: 'New Category', icon: 'Icon' };\n```\n\n## Search Index Update\n\nAfter adding prompts:\n```bash\nnpm run gen:search\n```\n\n**Ready to add or update prompts. What prompt are we working on?**", "variables": [], "notes": "Add, update, and organize prompts in the FrankX.AI Prompt Library", "metadata": { "source": { "provider": "github-claude-commands", "repository": "https://github.com/frankxai/frankx.ai-vercel-website", "path": ".claude/commands/frankx-ai-prompts.md", "ref": "f47422d3a97262dab0f98e8e89f2f791351017e8", "url": "https://github.com/frankxai/frankx.ai-vercel-website/blob/f47422d3a97262dab0f98e8e89f2f791351017e8/.claude/commands/frankx-ai-prompts.md", "key": "frankxai/frankx.ai-vercel-website/.claude/commands/frankx-ai-prompts.md" } } }

frankx-ai-prompts

About

FrankX.AI Prompt Library Management

Current Library Stats

Adding a New Prompt

Step 1: Determine Category

Step 2: Prompt Structure

Step 3: Quality Checklist

Step 4: Category Balance

Prompt Ideas from FrankX Content

From Intelligence Atlas:

From Vibe OS:

Updating Existing Prompts

Bulk Operations

Search Index Update

Install frankx-ai-prompts and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Use it

Manifest

Reviews