Creating and running AI-assistant skills — with a library in your storage

Describe a skill in chat — AI formats it into an instruction and saves it to a library in GitLab; when the skill is needed, the assistant finds it there and runs it.

  • A skill is created out of an ordinary conversation
  • All skills live in one library in GitLab
  • A finished skill runs on demand
  • No need to re-explain the same thing to the assistant every time
MCP Client Tool
Set this upScriptera

How it works

To pass leads from GitLab to MCP Client Tool automatically, use a ready-made scenario — no manual copying, no lost leads.

  1. Starts when: When chat message received
  2. Then: AI Conversational Agent
  3. Check: Proceed?
  4. If yes: Claude Skills Creator Agent
  5. Then: Extract Skill MD
  6. Then: SKILL.md Parser
  7. Then: Create a Skill
  8. Then: Skill created
  9. If no: More info
  10. Then: Simple Memory
  11. Then: Context7
  12. Then: Simple Memory1
  13. Then: OpenAI Chat Model
  14. Then: Anthropic Chat Model2
  15. Then: Simple Memory3
  16. Then: Structured Output Parser
  17. Then: List Skills
  18. Then: Skills Agent
  19. Then: Respond
  20. Then: Set chatbot vars
  21. Then: List Files
  22. Then: Get File from Skill
  23. Then: Anthropic Chat Model
  24. Then: Anthropic Chat Model1

You can launch this GitLab + MCP Client Tool integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.

Similar scenarios