The MCP Discovery
Transparency Index
Independent assessment of platform transparency in MCP server discovery
PREVIEW MODE
Currently, no platforms have implemented push-style MCP recommendations yet. This is a preview of how the index will look when scoring begins (Q2 2026).
Learn about push recommendations →Platform Transparency Scores
ChatGPT
OpenAI
Push-style MCP recommendations not yet implemented
Last updated: Jan 22, 2026
Claude
Anthropic
Push-style MCP recommendations not yet implemented
Last updated: Jan 22, 2026
Gemini
Push-style MCP recommendations not yet implemented
Last updated: Jan 22, 2026
Copilot
GitHub / Microsoft
Push-style MCP recommendations not yet implemented
Last updated: Jan 22, 2026
How We Score
Standards Adoption
Measuring adoption of community-proposed quality signals
- Uses proposed quality signals specification20 pts
- Publishes servers' quality scores10 pts
Process Transparency
Measuring openness of recommendation processes
- Documents weighting philosophy25 pts
- Explains inclusion/exclusion criteria20 pts
- Communicates algorithm changes15 pts
- Provides developer feedback10 pts
Total Score: 100 points • Letter Grades: A (90-100), B (80-89), C (70-79), D (60-69), F (<60)
FAQ
Why are all platforms greyed out?
No major platform has implemented push-style MCP recommendations yet. We're monitoring development and will start scoring as soon as platforms begin making proactive server suggestions.
When will you start scoring?
As soon as platforms implement push recommendations. We expect this to happen in Q2-Q3 2026 based on current development timelines.
What are "push-style recommendations"?
When an agent proactively suggests MCP servers based on user intent, rather than requiring users to manually search and install. This is where transparency becomes critical.
Get Notified When Scoring Begins
We'll email you when the first platform implements push recommendations and we publish our initial transparency assessments.