synthesize_feedbackLABSCO SUMMARY
Feedback read for you, with the receipts kept.
The output is built to be argued with: every cluster carries an impact score and links back to the individual issues and comments behind it, so a ranking can be checked rather than trusted.
What it is
An MCP server that collects customer feedback from GitHub Issues, Hacker News and Apple App Store reviews, runs it through a multi-pass LLM pipeline, and returns ranked pain clusters with impact scores, evidence links and suggested actions. Calls are paid per call in USDC over x402 rather than by subscription.
What you get
- Multi-source synthesis into ranked pain clusters, each with evidence links - the priciest of the four calls
- Pain points extracted from a single source
- Full-text search across cached feedback items - the cheapest of the four
- Sentiment over time across sources
Requirements
No signup, but you do need a wallet: an EVM private key in EVM_PRIVATE_KEY, funded with USDC on Base mainnet, since each call is paid as it runs. Install with pip and put the key in your client config. The sources are GitHub Issues, Hacker News and Apple App Store reviews - anything else has to reach you another way.
Setup effort
One command plus credentials — pip install feedback-synthesis-mcp, then supply credentials
USE CASES
01
Deciding what to build next quarter
Synthesise across sources and read the clusters in rank order, each with its evidence attached.
02
Finding out why the App Store rating slipped
Track the sentiment line, then pull the pain points from that source alone.
03
Checking whether a complaint is one person or a pattern
Search the cached items for the phrase and count what comes back.
04
Preparing a roadmap argument you can defend
Take the clusters with their impact scores and evidence links into the meeting.
05
Watching sentiment after a release
The time series covers all three sources, so a dip shows up wherever it starts.
06
Reading a repository's issue backlog without opening it
Single-source extraction is the cheap call, at two cents, for a first look.
4 tools
Listed in project docs. Your client may show a different list.
get_pain_pointssearch_feedbackget_sentiment_trendsBEFORE YOU INSTALL
What to know first
Tools you get
4 tools named in the project's documentation.
Requires
- feedback-synthesis-mcp on PATH
- EVM_PRIVATE_KEY in your environment
Where it runs
Both documented clients run the server on your machine.
INSTALLATION
Connect your client
Server setup
pip
TERMINAL
pip install feedback-synthesis-mcpHosted endpoint
ENDPOINT URL
https://feedback-synthesis-mcp-production.up.railway.app/mcp/Add this hosted endpoint to your MCP client using the credentials listed under Requirements.
MCP Clients
Claude Desktop, Cursor
JSON
{
"mcpServers": {
"feedback-synthesis-mcp": {
"command": "feedback-synthesis-mcp",
"env": {
"EVM_PRIVATE_KEY": "<YOUR_EVM_PRIVATE_KEY>"
}
}
}
}
