A
licenseA
qualityB
maintenanceMCP server for persistent, compounding memory that automatically captures corrections and insights across AI sessions, enabling agents to learn and improve over time.
5
371
MIT
mortgage-qa-memory
by Xavius123
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mortgage-qa-memoryRemember: LE generation flaky on Chrome, retry 3 times"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Mortgage QA Memory MCP
Status: working POC/MVP, now pivoting toward a governed knowledge-graph memory core. See PLAN.md for the full history (v1 design → v2 build → v3 roadmap) — start there.
A governed MCP memory service for Playwright QA automation, with tiered retention and mortgage compliance audit, adapted from DoorDash's and Salesforce's agentic memory architectures.
See AGENTS.md for agent working rules.
Who this is for
Platform / QA engineers building internal AI tooling on Cursor, Gemini gateway, KB MCP, and Azure MCP
Mortgage technology teams that need QA intelligence without creating a second store of loan data or NPI
Teams evaluating Playwright MCP + a QA memory expander they control end-to-end
Related MCP server: Memory Crystal MCP Server
Quickstart
npm install
npm run typecheck
npm test
npm run seed:demo
npm run smoke # QA-domain MCP tools
npm run smoke:graph # governed knowledge-graph MCP toolsThen point Cursor at cursor/mcp.json and add the
mortgage-qa-triage skill.
Layout
Path | Contents |
| The one enforced policy: retention, deny patterns, RBAC, write tiers |
| The runnable monorepo — see PLAN.md v2 |
| Curated (Tier 2) mortgage journey with TRID checkpoints |
| LL-2026-04 AI tool inventory |
| Cursor MCP config + triage skill |
| QA agent definitions (AC explorer, testcase writer, ADO publisher, automation generator, QA assistant) |
| Golden CI-failure set for flake-classification accuracy |
| Synthetic loan data used by tests |
Related external references
DoorDash Ask DoorDash / InfoQ summary — agentic memory + MCP + eval at scale
Playwright MCP docs — browser automation via MCP
flakiness-knowledge-graph-mcp — reporter + SQLite + MCP pattern
Fannie Mae LL-2026-04 — AI governance for seller/servicers (effective Aug 6, 2026)
Blend Autopilot MCP — lending MCP reference architecture
Next steps
See PLAN.md v3 for the live roadmap and open questions for the QA team.
This server cannot be deployed
Maintenance
ActivityMaintained
ResponsivenessSyncing
Related MCP Connectors
- memnodeOAuthdev.memnode
Persistent, inspectable memory for AI agents with lineage, correction, and a hosted MCP endpoint.
MCP-native Trust Infrastructure for AI Agents. Persistent encrypted memory with Trust Quotient.
Shared, governed long-term memory for AI agents across tools and sessions via MCP and REST.
Compliance frameworks (SOC 2, ISO 27001, CMMC, NIST, more) delivered to AI agents as MCP tools.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to maintain persistent memory across sessions by capturing conversations, extracting durable knowledge, and injecting relevant context, supporting various MCP-compatible platforms.12MIT
- AlicenseBqualityCmaintenanceBrowser automation MCP server with persistent memory for AI assistants, enabling automated web testing and workflow replay with self-healing selectors.543MIT
- AlicenseNot gradedqualityCmaintenanceEnables persistent memory for AI agents, combining episodic and semantic memory with LLM reasoning, accessible via MCP.2MIT
