Three pains every sales and BD team hits weekly. Each one is what your reps actually complain about, not what a feature page would call them.
Regression suites eat QA hours release after release, on both a web app and its mobile companion.
Every pipeline below is a shape you wire on the canvas using the crew and tools further down. Not a feature we ship for you, a pattern you configure.
Melaya Browser Control drives Chrome, Edge, or Brave through a checkout or onboarding flow, reading the real page structure rather than fixed pixel coordinates, so a run survives a small CSS change. FrontendEngineer walks the flow step by step and every consequential click pauses for approval.
Melaya Device Control walks the actual app on a paired Android phone: install, log in, tap through, one screenshot per step. DevOpsEngineer schedules the run per build and TechLead reads the evidence for a pass or fail call.
A QA lead who already lives in Claude Code or Cursor connects to api.melaya.org/mcp (OAuth 2.1) and grants melaya:browser (8,711 tools) or melaya:phone (8,711 tools) to run the same regression suite without opening the Melaya app.
Ahead of a release, the full tech_team crew reviews the diff for security, performance, and accessibility gaps in one sequential pass, SecurityAuditor and PerformanceExpert first, TechLead's verdict last.
Every failed step becomes a Jira or Linear issue with the screenshot and the step log attached, so a tester never retypes a repro by hand.
Real personas from the tech_team crew. Each ships with a tuned system prompt and a default tool allowlist. Swap models per persona on the canvas.
Owns the golden paths and signs off the pass/fail criteria.
Walks each client app screen by screen on real hardware.
Schedules parallel suites and ships the per-run reports.
Keeps demo runs presentable and flags visual regressions.
Every tool below is a real shared tool from the Melaya bundle. Allowlist per agent; HITL-gate the writes; revoke any of them in one click.
Replays demos and regression paths on physical phones, step by step.
Files failures and reports where the team already works.
Turns runs into client-ready reports.
Reads specs and logs alongside the device runs.
Every pipeline ships with three layers of knowledge access. Mix and match per agent on the canvas. No shared vector space with another tenant, no surprise reads, no opaque retrieval.
includeContextPer-pipeline documents appended to specific agents' input on every run. The ICP brief, playbook, pricing sheet, or won-deal email corpus. Whatever needs to be there before the agent thinks. You pick which personas get which docs.
rag_retrieveA scoped tool granted per-agent. When the agent decides it needs more depth, it queries the workflow's vector store on demand. Same knowledge base as Static context, accessed only when the model asks for it.
pipeline_memoryPipeline-level state that carries from one run to the next. Yesterday's research is in scope for today's follow-up. The crew remembers what it already prospected, what got approved, what was sent. The audit log is the second-order knowledge base.
Yes. Browser Control drives the web flow in a real Chrome, Edge, or Brave session and Device Control drives the Android app on a paired phone; both feed the same tech_team crew for the pass/fail verdict.
Real page structure. Agents read the page through browser_get_screen_tree and click by element reference, so a run survives a CSS-only change that would break a coordinate-based script.
No. Device Control drives the app through the screen the way a user does, with no SDK and no test-framework integration required.
Yes, through the Melaya MCP server at api.melaya.org/mcp. Grant melaya:browser or melaya:phone to Claude Code, Cursor, ChatGPT, or Le Chat and trigger runs from wherever the team already works.
The run pauses, captures a screenshot and step log, and can file a Jira or Linear issue automatically, with a recording and a pass/fail report per run.
Yes. The 10-persona tech_team crew runs a security, performance, and accessibility pass on the diff before the live regression run, with TechLead synthesizing a verdict.
Sandbox tier is free with no card. Join the waitlist and we will email you the moment a slot opens.
← Back to every use case