Installable skillMIT

Feature Flags & Progressive Delivery

An MCP server that provides on-demand skill loading for AI coding assistants.

BY modbenderUPDATED Oct 30, 2026
INTENDED FOR✳Claude Code>_Codex↗Cursor✦GeminiInstall in a listed agent.

instructions.md

---
name: feature-flags-progressive-delivery-354481adcefc
description: "An MCP server that provides on-demand skill loading for AI coding assistants."
---

# Feature Flags & Progressive Delivery

## Feature Flag Implementation

```typescript
// Simple feature flag service
const flags = new Map<string, boolean>()

export const featureFlags = {
  isEnabled(key: string): boolean {
    return flags.get(key) ?? false
  },
  
  enable(key: string) {
    flags.set(key, true)
  },
  
  disable(key: string) {
    flags.set(key, false)
  },
}

// Usage
if (featureFlags.isEnabled('new-checkout')) {
  return <NewCheckout />
}
return <OldCheckout />
```

## A/B Testing

```typescript
// Simple A/B bucket
function getBucket(userId: string, experiment: string): 'a' | 'b' {
  const hash = hash(`${userId}:${experiment}`)
  return hash % 2 === 0 ? 'a' : 'b'
}

// Usage
const bucket = getBucket(userId, 'checkout-button-color')
if (bucket === 'a') {
  return <Button color="blue" />
}
return <Button color="green" />
```

## Kill Switch

```typescript
// Emergency disable
app.get('/feature/:key', (req, res) => {
  if (req.query.disable === 'true') {
    featureFlags.disable(req.params.key)
    return res.json({ status: 'disabled' })
  }
  return res.json({ enabled: featureFlags.isEnabled(req.params.key) })
})
```

## Checklist

- [ ] Choose feature flag tool (LaunchDarkly, Unleash, custom)
- [ ] Implement flag evaluation
- [ ] Add user targeting
- [ ] Set up A/B experiments
- [ ] Create kill switches
- [ ] Plan rollout percentage
COMPATIBILITY & REFERENCES

Platform guidance

Imported from the listed AgentSkill.sh category. Complete supplied package files and notices are retained. Required external tools and credentials are not installed. No bundled code was executed during import. Individual agent task behavior has not been tested.