QED
About
Scores your writing against quality profiles — per-trait breakdown, deterministic, sub-second. No auth required.
Details
- Author
- onebit0fme
- Downloads
- 266
- Categories
- Other, Knowledge Base, Developer Tools
Jump to
- Deterministic scoring – same content, same profile, same result every time.
- Multiple quality profiles for different document types.
- Per-trait breakdown showing exact weaknesses.
- No LLM at inference – scores work in CI, scripts, or interactively.
- No auth, no signup required.
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
QEDCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Add QED to Claude Code via claude mcp add --transport http qed https://qed.systems/mcp. Then use the commands /qed:score, /qed:improve, or /qed:profile directly in Claude Code. No signup or authentication is required.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"qed": {
"qed": {
"type": "http",
"url": "https://qed.systems/mcp"
}
}
}
}McpServers
{
"qed": {
"type": "http",
"url": "https://qed.systems/mcp"
}
}QED scores text against quality profiles built from examples. Point it at a README, a landing page, or a technical doc — it returns a composite score plus a per-trait breakdown showing exactly what's weak and by how much.
Available profiles:
- qed.compelling-readme — hook speed, problem clarity, quickstart friction, credibility
- qed.technical-writing — narrative structure, grounded motivation, honest specificity
- qed.developer-landing-page — time to understanding, path clarity, show-don't-tell
- qed.puns — for when correctness matters less than commitment to the bit
Why deterministic matters: same content, same profile, same score every time. No LLM at inference — scores work in CI, in scripts, or interactively without nondeterminism.
No auth, no signup. Add to Claude Code:
claude mcp add --transport http qed https://qed.systems/mcp
Then use /qed:score, /qed:improve, or /qed:profile directly in Claude Code.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
