DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to rank files by code smells before a review with Supercov

Score a repository's code quality and see the weakest files and the smells that fired in each.

Short answer: Export your TypeSafe AI key with export TYPESAFE_API_KEY=... using a key from https://typesafe.ai; then run npx supercov quality to score the repository and list the weakest files; run npx supercov quality gaps to narrow the output to the smells that fired. All steps.

Credentials. Quality scoring needs a TypeSafe AI API key that you export into your shell, so keep it out of committed files and shared logs.

Who it is forA developer or tech lead who wants to know where a codebase needs cleanup before planning a refactor.

The problemWithout a score, it is hard to tell which files are worst or which specific smells, such as long methods or deep nesting, are driving the problem.

The outcomeA ranked list of weak files with the smells that fired, plus a per-file reading you can check against the source.

What you need first

  • A project with a complete test command the repository already runs, such as npm test, pytest or cargo test
  • Node.js so npx can fetch Supercov, or one of the other install routes in the README
  • A TypeSafe AI API key from https://typesafe.ai, only for quality and security scoring; coverage needs no account

Steps

  1. Export your TypeSafe AI key with export TYPESAFE_API_KEY=... using a key from https://typesafe.ai.
  2. Run npx supercov quality to score the repository and list the weakest files.
  3. Run npx supercov quality gaps to narrow the output to the smells that fired.
  4. Run npx supercov quality file <path> to read one file in full, using the path from the weakest list. (inferred, check the README)

Prompt for your agent

In the supercov repo's target project at <project path>, run `npx supercov quality` with TYPESAFE_API_KEY set, then `npx supercov quality file <path>` for the weakest file. Report the score, the smells that fired, and confirm each smell against the lines of that file.

Caveat. Scoring costs about a cent per megabyte of source, and Jev reads your source files to answer the questions, so it is not a fully local check.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for supercorp-ai/supercov and "Rank files by code smells before a review"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer