Agent use cases
Cited playbooks: what an agent can do with the servers in the directory.
76 use cases
Every playbook cites a vendor page and was reviewed before publishing.
| Use case | Category | Effect | Open |
|---|---|---|---|
Publish build status to iOS widgetsDevelopers see real-time CI/CD build status, queue depth, or job progress directly on their iPhone home screen widgets and Lock Screen Live Activities without opening an app. | Software development | Changes data in your accounts | |
Publish static site preview for 48 hoursA public HTTPS link to a temporary static site preview is generated, valid for 48 hours, allowing stakeholders to review HTML/CSS/JS content without permanent hosting. | Software development | Changes data in your accounts | |
Publish static site preview via AIfra MCPA temporary public URL for a static site preview is generated and returned, valid for 48 hours, allowing immediate sharing of web assets without permanent hosting. | Software development | Changes data in your accounts | |
Query Codebase Knowledge GraphThe agent returns structurally-aware answers about the codebase by querying the Knowledge Graph, providing deeper insights than simple text search. | Software development | Read-only | |
Query compliance overlay for delivery tasksThe agent retrieves specific, cited compliance guidance for software delivery questions, ensuring the output aligns with the customer's subscribed standards without hallucinating framework requirements. | Software development | Read-only | |
Recommend search provider for critical tasksThe agent receives a primary search provider, a fallback provider, and a recommendation ID to ensure reliable search execution without hardcoding providers. | Software development | Read-only | |
Record a new browser test via natural languageA new, editable AIVA test is created in the workspace, covering the specified user flow, ready for review, execution, and healing. | Software development | Changes data in your accounts | |
Register and publish a verified .agt agent identityThe agent's name is registered on-chain, its manifest is signed and hosted, and all endpoints are verified as live, establishing a trusted, verifiable identity for the agent. | Software development | Moves money or places orders | |
Report search execution outcome to AOPXThe agent records the real-world success or failure of a search operation, contributing to the provider's production evidence ledger for future routing decisions. | Software development | Changes data in your accounts | |
Resolve technical issues via agent networkThe agent receives a rated, high-quality solution from another AI agent to resolve a specific technical error or implementation challenge. | Software development | Read-only | |
Retrieve cached coding answers via AIephThe agent retrieves previously cached answers for identical or semantically similar coding questions, reducing API latency and costs by reusing existing knowledge instead of regenerating responses. | Software development | Read-only | |
Review Pull Requests with Real DocumentationGitHub Copilot code reviews cite actual project documentation instead of hallucinating, improving review accuracy and developer trust. | Software development | Read-only |
Browse by category
Where these playbooks come from
Each playbook is drawn from the vendor page it cites and reviewed by Koskamo before it is published. Koskamo does not run these servers or your agent: check what each tool will do before you let it act on your accounts.
Koskamo reads the example, tutorial, and guide pages that vendors publish for their own MCP servers. A playbook is only drafted when the page says the server can do it, and the supporting quote is shown on every playbook.