Click on "Install 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., "@Gemini Upgrade QA MCP Test new Gemini model for regressions against current version."
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.
Catch Gemini model upgrade regressions before they reach customers.
Gemini Upgrade QA is a paid remote MCP for Gemini upgrade evals, prompt regression checks, model output diffs, blocking rules, and eval receipts.
This is a public documentation project for Gemini Upgrade QA MCP. The structure is modeled after the public documentation pattern used by MiroFish: a short front door, a clear reading order, practical guides, reference pages, and public-safe architecture notes.
Support: support@aigeamy.com
Related MCP server: hallumark
Endpoint: https://geminiupgradeqa.clauxel.com/mcp
Server card: https://geminiupgradeqa.clauxel.com/server-card.json
Registry name: com.clauxel.geminiupgradeqa/geminiupgradeqa-mcp
Tools: run_gemini_upgrade_eval, compare_prompt_outputs, detect_model_regression, issue_upgrade_receipt, export_eval_audit
AI platform teams, prompt owners, QA leads, and release engineers.
upgrade eval runner
prompt output comparison
regression detection
blocking rules
eval receipt export
This repository does not contain production source code, credentials, payment configuration, Cloudflare configuration, customer records, private analytics, or local machine paths.
This server cannot be installed
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
MCP-native AI evaluation: rubric audits, eval suites, and proof reports for AI/LLM output.
Synthetic checks, nightly regression replay and model-drift alerts for AI agents
Create, validate and audit llms.txt, incl. the Lighthouse Agentic Browsing check.
Check AI work against requirements and return structured verdicts, findings, and repair steps.
