Stop your AI agent from rebuilding what already exists. An opinionated MCP server that checks GitHub, npm and Python repos for maintained alternatives before you scaffold.
View on GitHub ↗Homepage ↗Your idea probably already exists. Find out before you build it, not after.
Here is how it usually goes. You have an idea. You ask a frontier model what it thinks, and it tells you the space is wide open and people would love it. So you tell your agent to build it, and it writes thousands of lines for something that already exists.
That happened to me. I kept hitting the same failing Android CI build, waiting for a re-run, hitting it again. I thought: someone should be able to pause the build and get a shell inside it. I asked an AI, it told me this was a genuinely good idea, and I built fermata.
Then I found action-tmate. 3,566 stars. Does exactly that. Has for years.
This server is the check I wish I'd run. Ask your agent to build something and it looks first:
1. action-tmate (3,566★, updated today) — "Debug your GitHub Actions via SSH by using tmate to get access to the runner system itself."
Actively maintained and widely used. Want to try it, or is there something yours would do differently?
Sometimes the answer really is "mine is different, keep going." Sometimes it saves you a weekend. Either way you find out in ten seconds instead of after.
Why bother: fewer wasted hours, fewer abandoned projects nobody uses, and less energy burned generating code that already exists. Jenna Pederson put the general case well in You Can Build It, But Should You? — AI removed the friction that used to force you to ask whether something is worth building at all.
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/aradar46--reuse-before-generate.html)
Multi-agent orchestration for Claude Code: MCP server + skill that keeps Claude as lead engineer while delegating backend work to OpenAI Codex and frontend/design work to Kimi
azimov777/casefileThe task tracker your AI agents keep for each other: every task carries a case file, so the next agent picks up where the last one stopped. MCP server + live board. Self-hosted, one-line install.
00200200/usagetrimKeep the signal. Cut the noise. Local CLI + MCP — fold verbose tool output for Claude Code, Codex, Cursor & Desktop.
Riztazz/claude-io-guardA Claude Code plugin that checks every file and shell call an AI agent makes, fixes what it safely can, and tells the agent how to fix the rest. Windows and macOS.
dontpayfull/AInotateAnnotated screenshots for AI agents: Skitch-style arrows, numbered steps, boxes, labels, solid redaction
menot-you/n-memoryMemory your agent can trust — every answer carries its source; when it doesn't know, it says so. Hermetic, local, MCP.
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.