Flamingo Raises $4.5M Seed Round

Try for Free
Back to Life at Flamingo

How I Use Skills and Claude Design 🦩

AIAI INTEGRATIONAUTOMATIONBEST PRACTICESGUIDEMARKETINGWORKFLOW AUTOMATION

Sep 2, 2026

Session

Marketing

Discipline

Beginner

Level

Kristina Shkriabina

Kristina Shkriabina

Content Marketing Lead

Kristina presented a comprehensive overview of using skills and cloud design systems for workflow automation. She demonstrated how skills function as reusable prompts that reduce repetitive work, emphasized the critical importance of detailed descriptions over naming, and shared her personal automation setup using 13 skills and multiple cloud code routines. Key takeaway: skills should be continuously improved and adapted to specific workflows, with proper structure and guardrails to ensure quality output.

Tools Used and What For
1

  • Setup

    Kristina's setup: 13 skills and 2+ plugins, connected through Claude Code as hub
    Uses 10+ MCPs (model context protocols) and GitHub repositories for skill storage
    Connected to Supabase for automation flows without manual posting
    Synthetic sessions inside Claude Code can replicate browser experience for tools without API/MCP support

Reusable Takeaways
1

  • Skill Creation Best Practices

    Description should contain all prompt components: context, task, rules, refusals, error handling
    Include examples rather than generic instructions to improve output quality
    Put most important information first to speed up processing
    Skills can range from simple single files to complex with hundreds of supporting documents
    Claude loads only relevant documents to avoid wasting tokens

Failures and Gotchas
1

  • Common Skill Mistakes

    Skill not firing or wrong skill firing: usually due to confusing descriptions
    Similar descriptions across multiple skills causes wrong selection
    Outdated information hardcoded in skill body instead of using flow rules
    Process changes without updating corresponding skills
    Need to set up evals to regularly check skill performance

Take This With You

Kristina Shkriabina

Kristina Shkriabina

Content Marketing Lead

Ohayo! I run content, SEO, social, and community at Flamingo. Before IT, I worked as a correspondent for Ukraine's Public Broadcasting Company and have a Master's in journalism.

Frequently Asked Questions

About OpenFrame

OpenFrame isn't built to plug into your stack. It replaces it. Instead of duct-taping a dozen tools together (RMM, MDM, SIEM, patching, remote access, each its own login and bill), we bundle it into one unified platform: RMM, MDM, monitoring, automation, remote access, patch management, security monitoring, and ticketing, plus built-in AI copilots. So "does it integrate with X?" usually means: you won't need X anymore.
Most platforms give you one piece and expect you to bolt the rest on. OpenFrame unifies the whole stack in one place, with AI copilots built in. Fewer logins, fewer bills, less duct tape.
In the cloud, on US soil. Your data stays stateside.
Both. It's built for MSPs and MSSPs alike.

MSP AI Agents

Yes. In production MSP shops today, 10% to 25% of tickets close before a human opens them. Thread alone has processed 173 million tickets across 750-plus MSP partners at 96% triage accuracy, handing back 490,000-plus technician hours. Agents own the low-risk, high-volume work (password resets, MFA enrollment, known installs, onboarding and offboarding) and flag anything that touches production data or needs judgment for a human to take.
On a five-person desk, reported deployments show $78,000 to $130,000 in annual direct labor savings, roughly 30% fewer escalations, and 15% to 20% better SLA compliance. Broader MSP adoption data adds ticket handling time cut by 45% and five to 12 points of margin, all from reclaimed capacity rather than headcount cuts.