Laravel performance, in your terminal

Find slow routes and N+1 queries
without leaving your IDE.

Profile route latency, detect N+1 loops, and jump to the offending line in your IDE. Zero telemetry.

Get started → Read the docs

Tests Passing PHPStan Level 5 Laravel 12–13 MIT License

Pinpoint/Diagnostics

$php artisan pinpoint:report

Route

P95

Avg

Status

N+1

Measured runtime overhead

+0.29 ms mean overhead. Telemetry flushes asynchronously after your HTTP response is dispatched. View benchmark code

See the fix

GET /api/orders · OrderController@index

-93.7% latency

Unoptimized (N+1) Critical

1,420ms

42 queries · 14× lazy-loaded relations

Resolved via Pinpoint Optimized

89ms

3 queries · Order::with('items')

Workload: 1,000 Order records iterated on local PHP 8.3 / MySQL. Found in seconds via pinpoint:report

01 / Core Capabilities

Runtime visibility without the runtime tax.

Pinpoint connects runtime performance directly to the source lines responsible for them.

01 Detect

N+1 and query anomalies

Find repeated relationship queries before they become production bottlenecks. Uses Laravel's native lazy-loading signals, with SQL fingerprinting as a fallback.

Eloquent signal SQL fingerprint

Query pattern 14× Duplicate

Order → items

SELECT * FROM `items` WHERE `order_id` = ?

02 Measure

Route latency & percentile rollups

Understand request performance with p95/avg latency and sample counts. Rollups compute on-demand via a single indexed SQL pass, or batched via pinpoint:aggregate.

Overhead methodology (0.29ms mean)

Request Diagnostics

P95

1.42s

Avg

890ms

Queries

42

03 Locate

Source-level context & IDE jump

Ctrl/⌘-click any route or caller in your terminal to jump directly to the controller method and offending line in VS Code or PhpStorm. Zero telemetry, 100% local.

vscode:// · phpstorm:// terminal links

Source Origin

OrderController.php:84

Suggested: Order::with('items')

pinpoint:report --tier=critical

api.orders.index

N+1 Violation Detected

Relationship: Order → items 14 executions

SQL: SELECT * FROM `items` WHERE `order_id` = ?

Trigger: Eloquent Lazy Load detected during iteration

02 / CI Regression Gate

Fail the merge before an N+1 reaches production.

Run Pinpoint as an automated CI regression gate in GitHub Actions. Automatically exit with code 1 when query budgets or N+1 violations occur.

GitHub Actions • CI Terminal

Exit 1 • Check Blocked

$ php artisan pinpoint:check --fail-on-n1

✗ FAIL api.orders.index

1x N+1 · Order::items()

Trigger: app/Http/Controllers/OrderController.php:34

Process completed with exit code 1. 0.14s total

03 / Tested in the Wild

Real output from a real production codebase.

An unedited CLI report profiling a 20-route production application—spotlighting latency bottlenecks and N+1 query loops in milliseconds.

1 Critical Bottleneck

v1.historical-records.search clocked at 4,350ms, flagged instantly in the red tier.

5 N+1 Violations

Detected relationship loops with repeat counts up to x11 queries per single HTTP request.

Instant Locate Block

Clickable route labels and caller breadcrumbs jump straight into your IDE to resolve the root cause.

04 / Comparison

Built around the Laravel runtime.

Actionable terminal diagnostics and CI regression gates without the overhead of external SaaS observability stacks.

Swipe to compare →

	Pinpoint	Telescope	Pulse	Laravel Slower
Primary focus	Route & N+1 latency	General debug assistant	Server health & queues	Slow queries + AI fixes
CI regression gate	Yes — blocks PR before merge	—	—	—
Fix guidance	Exact eager-load fix (deterministic, free)	Raw query view only	Aggregate metrics only	AI-generated: indexes, rewrites, eager-loads
N+1 query precision	Exact relation & model (Eloquent hook)	Manual query count logs	Aggregate slow query list	Duration-threshold + fingerprint grouping
Source line traceback	Exact file and line	Backtrace in web modal	—	Exact file and line
Execution environment	CLI and native IDE	Local web dashboard	Central web dashboard	Self-hosted dashboard (/slower)
Telemetry residency	Local machine only	In-app database	Redis / app database	Local by default; SQL/schema to LLM on demand
Request overhead	0.29ms mean (deferred)	High (inline recording)	Low (sampled)	Threshold-gated; sampling + circuit breaker

Ready to profile your routes locally?

Zero telemetry, zero dashboard overhead. Up and running in 60 seconds.

Get started → View source

05 / Documentation

Get started in minutes.

Install, migrate, run your app. Then one command tells you what's slow.

01

Install as a dev dependency

$ composer require asimali/pinpoint --dev

02

Publish & run migrations

$ php artisan vendor:publish --tag="pinpoint-migrations" && php artisan migrate

03

Generate your first latency report

$ php artisan pinpoint:report

Complete Artisan Toolkit

Everything needed to profile locally, gate in CI, and manage retention.

$ php artisan pinpoint:report

Primary Diagnostic

Generate terminal summary tables with percentile rollups (p95/avg), N+1 repeat multipliers, clickable IDE links, and machine-readable JSON payloads.

Flags --tier=critical --route=<name> --since=30m --limit=10 --json --json-to=<file>

$ php artisan pinpoint:check

Evaluate query counts, durations, and N+1 violations as an automated CI regression gate, exiting with code 1 upon failure.

Flags --fail-on-n1 --max-queries --max-duration-ms --allow-empty --json

$ php artisan pinpoint:aggregate

Batch-compute historical hourly and daily percentile rollups in worker jobs with transaction-safe chunking.

Flags --force

$ php artisan pinpoint:prune

Delete request and query samples older than the retention window (default 7 days) via database cascade.

Flags --days=7 --force

$ php artisan pinpoint:reset

Truncate and purge all recorded telemetry data from the database with confirmation prompt.

Flags --force

What Pinpoint adds to each request

Recording runs after your response is sent via app()->terminating(). A 1,420ms route measures 1,420.29ms with Pinpoint on — 0ms added to TTFB.

0.29ms

Pinpoint