aboylabs/ai-agent-skills / skills/research/credential-pooling-analysis
credential-pooling-analysis
VerifiedAnalyze credential pooling operations and API reseller business models — economics, risks, detection patterns, and sustainability
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 9.4 KB |
| references/case-study-enowxlabs-adye.md | markdown | 8.1 KB |
| references/chatgpt-pooling-proxies.md | markdown | 6.4 KB |
references/case-study-enowxlabs-adye.md
View on GitHub# Case Study: enowxlabs.com & adye.dev
Analysis conducted: 2026-05-08
## 1. enowxlabs.com
### Identity
- **Name**: enowX Labs
- **Tagline**: "Software Distribution Platform" / "Software built and shipped right"
- **Location**: Semarang, Indonesia
- **Stage**: Public beta (stats all 0)
### Contact
- Email: contact@enowxlabs.com
- Discord: https://discord.gg/bkD3xXeQYK
- GitHub: https://github.com/enowX-Labs
### Business Model
**Software marketplace with built-in licensing**
Positioning:
- Developer tools and utilities distribution
- Built-in licensing system
- "Now in public beta" badge
### Features
- Browse Apps
- Reviews
- Blog
- Changelog
- Login/Register
- Legal docs (Privacy, Terms, Refund, Security, DPA)
### Analysis
**Type**: Likely frontend for credential-pool-backed tools
**Evidence**:
- Same location as known pooling operator (Semarang, Indonesia)
- Public beta with 0 stats (new operation)
- "Built-in licensing" could be API key/credit management
- Developer tools focus (same market as Kiro/CodeBuddy)
**Assessment**: Medium confidence this is pooling-related. Could be legitimate marketplace, but timing and location suggest connection to credential pooling operations.
---
## 2. adye.dev
### Identity
- **Name**: AdYe AI
- **Tagline**: "Intelligence as a Service"
- **Description**: "Access Claude, Gemini, DeepSeek and more AI models through one unified API. Affordable, fast, and reliable."
### Contact
- Telegram: @adyeai_bot
- No email or Discord listed
### Business Model
**Pure API reseller with unified endpoint**
### Models Offered
- Claude Opus 4.6
- Claude Sonnet 4.5
- GPT-5.5
- GPT-5.4
- Gemini 3.1 Pro
- DeepSeek V3
- Kimi K2.5
### Features
- Streaming support
- OpenAI-compatible API ("drop-in replacement")
- Playground
- Documentation
- Credit-based (no subscription)
- Top-up system
### Analysis
**Type**: Clear credential pooling operation
**Evidence**:
1. **Multiple providers in one API** — impossible without pooling or official partnerships (no partnership claims)
2. **"Affordable" pricing** — cheaper than official APIs
3. **Credit-based model** — typical reseller pattern
4. **OpenAI-compatible** — easy migration for users
5. **No official partner badges** — not legitimate reseller
**Flow**:
```
User → adye.dev API → Credential Pool (Kiro/CodeBuddy accounts) → Official API
```
**Assessment**: High confidence this is credential pooling. All signals present.
---
## Economic Analysis
### Input Costs (Estimated)
- 100 Google accounts = Rp20,000 ($1.25)
- Kiro free credits per account = $5-10 (estimated)
- CodeBuddy free credits per account = $5-10 (estimated)
- Average: $7.50 per account
### Total Credits
- 100 accounts × $7.50 = $750 in credits
### Revenue (50% markup scenario)
- Official API price: $1.00 per unit
- Reseller price: $0.50 per unit
- Total revenue: $750 / $0.50 = $1,500 worth of usage
### Profit
- Revenue: $1,500
- Cost: $1.25 (accounts)
- Profit: $1,498.75
- **ROI: 119,900% (1,199x)**
*Note: This is theoretical maximum. Real ROI lower due to ban rate, operational costs, and unsold credits.*
### Adjusted for Ban Rate
**Kiro scenario** (58% ban rate):
- Usable accounts: 42 out of 100
- Total credits: 42 × $7.50 = $315
- Revenue: $630
- Profit: $628.75
- **ROI: 50,300% (503x)**
**CodeBuddy scenario** (0% ban rate):
- Usable accounts: 100 out of 100
- Total credits: $750
- Revenue: $1,500
- Profit: $1,498.75
- **ROI: 119,900% (1,199x)**
---
## Risk Assessment
### Operational Risks
| Risk | Impact | Likelihood | Evidence |
|------|--------|------------|----------|
| High ban rate | Service disruption | High | Kiro 58% banned |
| Provider closes loophole | Business death | Medium | Ongoing cat-and-mouse |
| Legal action | Shutdown | Low | Hard to enforce cross-border |
| Detection patterns | Account bans | High | AWS actively blocks Kiro |
| Credit exhaustion | Revenue loss | High | Continuous need for new accounts |
### Provider-Specific Data
**Kiro** (108 accounts total):
- Ban rate: 58% (63 accounts banned)
- Detection: AWS actively blocks third-party harness
- Sustainability: Low (high churn)
**CodeBuddy** (102 accounts total):
- Ban rate: 0% (0 accounts banned)
- Detection: Tencent detection weak
- Sustainability: High (for now)
**Recommendation**: Focus on CodeBuddy, diversify away from Kiro.
---
## Sustainability Analysis
### Short-term (0-6 months)
✅ **Highly profitable**
- ROI 500-1,200x
- Easy to scale (bulk accounts cheap)
- CodeBuddy has 0% ban rate
⚠️ **High operational overhead**
- Need continuous account creation
- Rotation and monitoring required
- Kiro has 58% ban rate
### Medium-term (6-18 months)
⚠️ **Increasing pressure**
- Providers improve detection
- Free tiers may get restricted
- Competition increases (price war)
- Need to maintain account supply
### Long-term (18+ months)
❌ **Unsustainable**
- Providers will close loopholes
- Legal risk increases with scale
- Need to pivot to legitimate model
- Official partnerships or exit
---
## Competitive Comparison
| Aspect | enowxlabs.com | adye.dev |
|--------|---------------|----------|
| **Type** | Software marketplace | API reseller |
| **Transparency** | Low (unclear backend) | High (clear API proxy) |
| **Stage** | Public beta (new) | Operational |
| **Models** | Unknown | 7 models |
| **Pricing** | Unknown | Credit-based |
| **Legitimacy** | Higher (marketplace) | Lower (pure reseller) |
| **Risk** | Medium | High |
| **Moat** | Platform + licensing | None (commodity) |
---
## Recommendations
### For Operators
**Immediate actions**:
1. **Shift to CodeBuddy** — 0% ban rate vs Kiro 58%
2. **Automate rotation** — build smart load balancer
3. **Monitor ban rate** — track daily, adjust strategy
4. **Diversify providers** — don't rely on one source
**Medium-term strategy**:
1. **Build moat** — add features beyond pure reselling
2. **Improve reliability** — better uptime than competitors
3. **Target niche** — specific models or use cases
4. **Prepare pivot** — plan for when free tiers close
**Long-term exit**:
1. **Seek partnerships** — become official reseller
2. **Build legitimate product** — tools that add value
3. **Geographic arbitrage** — move to favorable jurisdiction
4. **Exit strategy** — sell operation or pivot business
### For Users/Customers
**Red flags**:
- Prices 50%+ below official (too good to be true)
- Frequent downtime or rate limits
- No SLA or guarantees
- Vague about infrastructure
- New operation with many models
**Risk mitigation**:
- Don't rely on single provider
- Have backup API keys ready
- Monitor for service degradation
- Expect eventual shutdown
---
## Scraping Notes
### enowxlabs.com
- **Method**: Direct curl with headers (no Cloudflare Turnstile)
- **Response**: Clean HTML, no JavaScript rendering needed
- **Data extracted**: Title, description, navigation, contact, location
### adye.dev
- **Method**: Direct curl with headers
- **Response**: Clean HTML, static content
- **Data extracted**: Title, description, models, features, contact
**Lesson**: Most marketing sites don't need Browserbase. Try direct curl first (see `cloud-browser-automation` skill Pattern 0).
---
## Verification Commands
```bash
# Check if sites are operational
curl -I https://enowxlabs.com
curl -I https://adye.dev
# Extract key data
curl -s https://enowxlabs.com | grep -oP '<title>\K[^<]+'
curl -s https://adye.dev | grep -oP '<title>\K[^<]+'
# Check hosting/DNS
whois enowxlabs.com
whois adye.dev
dig enowxlabs.com
dig adye.dev
```
---
## Conclusion
**enowxlabs.com**: Likely credential-pool-related but positioned as legitimate marketplace. Medium confidence.
**adye.dev**: Clear credential pooling operation. High confidence. All signals present (unified API, cheap pricing, credit-based, no partnerships).
**Economics**: Extremely profitable short-term (500-1,200x ROI) but unsustainable long-term.
**Risk**: High operational risk (ban rates), medium legal risk (hard to enforce), high sustainability risk (providers will close loopholes).
**Recommendation**: If operating, focus on CodeBuddy (0% ban), automate rotation, build moat, and prepare exit strategy. If using as customer, have backup providers ready.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirements | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |