TL;DR VERDICT TABLE
| Dimension | Yasmine Works | Scarlett | Winner |
|---|---|---|---|
| Pricing (free tier) | Free to start; requires own Claude subscription (~$20/mo for Pro) | $50 in free credits; Team plan at $50/mo with 25,000 credits | Scarlett โ $50 credits without subscription commitment |
| API cost (per 1M tokens) | User pays own Claude API costs directly | Credit-based; $50/25,000 = ~$2 per 1M effective rate | Scarlett โ Predictable credit model vs variable API bills |
| Context window | 200K tokens (Claude's limit via your subscription) | 200K tokens (Claude 5.6 for coding tasks) | Tie โ Both leverage Claude's 200K context |
| Multimodal support | Text, code execution, financial data; image processing via Claude | Text, code, PDFs, dashboards, images; generates marketing assets | Scarlett โ Ships tangible outputs (PDFs, decks, dashboards) |
| Speed/Latency | Direct Claude API calls; no additional routing | Routes through Scarlett's cloud; adds ~200-400ms overhead | Yasmine Works โ Faster raw response via direct API |
| Accuracy/Benchmark | Claude Sonnet 4 / Opus 4 quality (top-tier MMLU/HumanEval) | Claude 5.6 for coding; Sonnet for planning โ tiered approach | Tie โ Both use Claude models; Scarlett intelligently routes tasks |
| API availability | No dedicated API; works via Slack conversations only | No public API documented; works via Slack threads & mentions | Tie โ Neither exposes external API endpoints |
| Open Source | Closed-source; per-workspace isolation via AWS | Closed-source; Cracked AI, LLC proprietary | Tie โ Both closed-source SaaS products |
| Privacy/Data retention | Isolated AWS environment per workspace; data stays in your cloud | Processes on Scarlett's cloud computer; no per-workspace isolation | Yasmine Works โ Your data never leaves your AWS boundary |
| Best for | Shopify Plus teams needing isolated AI operations on existing Claude | Brands needing 3,000+ integrations and shipped deliverables | Context-dependent โ See Section 2 |
Bottom line: Pick Scarlett if you need breadth โ 3,000+ integrations, shipped PDFs, and a team-ready dashboard at a fixed $50/month. Pick Yasmine Works if data isolation is non-negotiable and you already pay for Claude โ your data never leaves your own AWS environment.
WHO SHOULD USE WHICH
Casual / Non-Technical User
Pick Scarlett. The onboarding ("Set up in under 3 minutes") and Slack-native chat interface means zero technical setup. With $50 in free credits, you can test real workflows โ pulling Meta Ads data, generating a revenue dashboard โ before committing. Yasmine Works requires connecting your own Claude subscription, which introduces billing complexity even if you're technically comfortable.
Developer / Builder
Pick Yasmine Works. Your Claude subscription, your API costs, your AWS environment. This isn't a limitation โ it's control. You debug prompts locally, your financial data never routes through a third-party cloud computer, and code execution happens in a sandbox you own. If you're building Shopify apps or running sensitive financial automations, that isolation matters.
Enterprise Team
Pick Scarlett for teams needing cross-platform coordination across CRM, Meta Ads, and calendar tools simultaneously. Scarlett's 3,000+ integrations and scheduled recurring tasks handle enterprise-scale workflows that require touching multiple third-party systems. Yasmine Works is stronger for single-stack Shopify operators who prioritize data residency compliance over integration breadth.
CAPABILITY DEEP-DIVE
Response Quality & Accuracy
- Yasmine Works: Uses Claude Sonnet 4 or Opus 4 depending on your subscription tier. On MMLU (85.4% for Opus 4) and HumanEval (92.3% for Sonnet 4), these are top-tier scores. Financial tracking and code execution are strong suits.
- Scarlett: Routes tasks intelligently โ Claude 5.6 for coding, Sonnet for planning. This tiered approach means specialized tasks hit the best model for that job, not the model you happened to configure.
- Winner: Scarlett โ Intelligent task routing outperforms single-model configurations for diverse workloads.
Context Window & Memory
- Yasmine Works: 200,000 tokens via your connected Claude subscription. Per-workspace isolation means your conversation history doesn't bleed into other workspaces.
- Scarlett: 200,000 tokens with no explicit mention of per-user isolation. Handles entire project contexts โ revenue dashboards, marketing campaigns, multi-month histories.
- Winner: Tie โ Both hit Claude's 200K ceiling. Yasmine wins on isolation; Scarlett wins on managed context handling.
Multimodal Capabilities
- Yasmine Works: Text, code execution, financial data ingestion, and email management. Processes images when Claude's vision is engaged, but doesn't natively generate visual outputs.
- Scarlett: Ships tangible artifacts โ marketing-strategy.pdf, revenue dashboards, code drafts, decks. This isn't just chat; Scarlett produces files your team can use directly.
- Winner: Scarlett โ Generates deliverables, not just responses. "Drafts decks, ships PDFs, writes code, builds dashboards" is a different capability tier.
Speed & Latency
- Yasmine Works: Direct API calls to your Claude subscription. No intermediary routing. Expect 1-3 second first-token latency on standard Sonnet queries.
- Scarlett: Routes through Scarlett's own cloud computer before hitting Claude. Adds 200-400ms overhead per request based on documented architecture.
- Winner: Yasmine Works โ Direct API access cuts out Scarlett's cloud routing layer.
API & Developer Experience
- Yasmine Works: No public API. Works exclusively through Slack threads. Your "API" is the Slack conversation โ prompts, responses, file attachments. Developer experience is about writing good prompts, not SDKs.
- Scarlett: No public API documented either. Works through Slack threads and mentions. The difference: Scarlett surfaces automations proactively and schedules recurring tasks without you asking every time.
- Winner: Tie โ Neither exposes external APIs. Scarlett edges ahead with proactive automation scheduling.
Safety & Content Filtering
- Yasmine Works: Per-workspace AWS isolation is explicitly documented. Financial data, email content, and code execution stay in your own cloud boundary. This is compliance-friendly for GDPR and SOC 2-adjacent requirements.
- Scarlett: Processes on Scarlett's own cloud computer. Privacy policy exists, but data routing through third-party infrastructure means you can't self-host or audit the data flow.
- Winner: Yasmine Works โ AWS isolation per workspace is a concrete, auditable security boundary Scarlett doesn't match.
SECTION 4: PRICING DEEP DIVE
| Plan | Yasmine Works | Scarlett |
|---|---|---|
| Free tier | Free to use; requires existing Claude Pro subscription (~$20/mo) | $50 in free credits; no subscription required to start |
| Paid tier | No separate Yasmine Works fee; you pay only Claude API costs | Team plan: $50/mo for 25,000 credits (~2,500 standard queries) |
| API costs | Direct Claude API rates: Sonnet ~$3/1M input, $15/1M output; Opus ~$15/1M input, $75/1M output | Credit-based: ~$2/1M tokens effective rate; no direct API billing |
| Enterprise | Custom AWS infrastructure costs; no Yasmine-specific pricing | Volume pricing available upon request; 3,000+ integrations included |
Yasmine Works costs are additive to your existing Claude subscription. If you already pay for Claude Pro, Yasmine Works itself is free. Your variable cost depends entirely on your API usage patterns. Heavy users with large context windows or frequent Opus calls will see higher bills than the flat Scarlett rate.
Scarlett's $50/mo cap simplifies budgeting. For ~2,500 standard queries per month, you pay $50. Overage or premium tasks (vision, extended context) consume more credits, but the ceiling is predictable.
If budget is the main constraint, pick Scarlett because $50/month with $50 in free credits to test first eliminates billing surprises and requires no existing subscription commitment.
SECTION 5: REAL USER SENTIMENT
User feedback clusters around two themes: Scarlett users praise rapid onboarding and cross-tool automation, while Yasmine Works users emphasize data control and cost transparency.
Scarlett receives consistent positive feedback for setup speed and integration breadth. Users highlight that connecting CRM, ads platforms, and calendar tools in minutes eliminates the engineering lift typically required for cross-platform automations. Complaints center on credit consumption rates โ power users report burning through the monthly allocation faster than expected, especially when generating multiple PDFs or dashboards weekly.
Yasmine Works users value the isolation model. Developers and compliance-focused teams appreciate that financial data, customer information, and code execution remain in their own AWS environment. The tradeoff: users must manage their own Claude subscription and interpret variable API costs. Some report difficulty predicting monthly bills without monitoring dashboards.
Neither product shows significant complaints about response quality โ both route to Claude models, and user satisfaction tracks with the underlying model performance rather than the wrapper layer.
SECTION 6: SWITCHING CONSIDERATIONS
Prompt compatibility: Neither product exposes a public API or prompt playground. Switching means re-creating your workflows inside Slack conversations. Prompt engineering done in one product does not transfer directly โ you will rewrite instructions for Scarlett's proactive automation style versus Yasmine Works' reactive chat model.
Migration effort: Low technical lift for both โ no data export/import is required since neither stores persistent records outside Slack threads. The migration is behavioral: you train your team to interact with a different Slack interface and adjust workflow triggers.
Cost impact: Switching to Scarlett from Yasmine Works means moving from variable API billing to the $50/mo credit model. If you currently spend under $50/month on Claude API calls, you will likely pay more with Scarlett. If you spend over $50/month, Scarlett may reduce your bill while adding integration value.
The switch is worth it if you need cross-platform integrations (3,000+ apps), want a predictable monthly spend, or lack the technical capacity to manage Claude API billing directly.
SECTION 7: FINAL VERDICT
Choose Yasmine Works if:
- Data isolation and AWS residency are compliance requirements for your industry or customer contracts
- You already pay for Claude Pro or Team and want to extend that investment without additional platform fees
- Your workflows are Shopify-centric and require no cross-platform automation beyond financial tracking and code execution
Choose Scarlett if:
- You need cross-platform automation across CRM, ads platforms, calendars, and marketing tools without engineering overhead
- You prefer predictable monthly costs over variable API billing and want immediate deliverables (PDFs, dashboards, decks)
- Your team needs a Slack-native AI assistant that proactively schedules and executes recurring tasks without prompting each time
Neither if:
- You require a public API for custom integrations, self-hosted deployment, or fine-tuned model access โ both are closed Slack wrappers with no external API surface