Hotspots

Find where your engineering attention has the highest expected value.

Most of the pain in a codebase comes from a small fraction of it — the same files that generate the most incidents, the slowest reviews, and the hardest bugs to track down. Hotspots finds that fraction before it costs you.

hotspots analyze src/

LRS   File                      Line  Function
12.4  src/api/billing.ts         142  processPlanUpgrade   critical
 9.8  src/auth/session.ts         67  validateSession      high
 3.2  src/utils/format.ts         12  formatDate           low

Each result is ranked by Local Risk Score (LRS) — a weighted combination of cyclomatic complexity, nesting depth, fan-out, and exit paths. High LRS means the function is structurally hard to reason about, test, and safely change.

---

Get Started

Quick Start →Set Up CI →CLI Reference →

---

What it gives you

An objective refactor list. Stop debating what to clean up. The highest-LRS functions in your codebase are the ones most likely to slow your next feature and hide your next bug.

CI enforcement. Block new critical-complexity functions before they merge. Delta mode shows exactly which functions got worse in a PR — and by how much.

Progress you can show. "We dropped from 31 critical functions to 18 this quarter" is a concrete metric. Hotspots makes that trackable without extra tooling.

Everything stays local. Analysis runs on your machine. No source code leaves, no account required, no data sent anywhere.

---

Supported Languages

TypeScript · JavaScript · Go · Python · Rust · Java · C

All languages produce the same metrics with consistent semantics. See Language Support.

---

Where this is going

Risk hotspots — structurally complex, frequently changed code — are the first category. Several more are in active research:

• Review Hotspots — changes that need senior eyes, not rubber stamps
• Test Hotspots — coverage gaps where CI misses real failures
• Ownership Hotspots — knowledge silos and review bottlenecks
• Impact Hotspots — code with outsized blast radius (auth, billing, schema contracts)

The goal is a continuous picture of where your engineering attention matters most — before something ships, not after it pages you.

---

For Contributors

Start with the Codebase Guide for the implementation map, then the Contributing Guide for dev workflow.