14 Laravel packages tagged “evals”
Most downloaded: pestphp/pest-plugin-evals, with 188,036 downloads and 5 GitHub stars. Ranked by total downloads — use the controls below to sort by stars, most recently updated, or name.
pestphp/pest-plugin-evals
A PestPHP plugin for evaluating LLM agents with LLM-as-judge, semantic similarity, and deterministic scorers
5
188,036
vbarbarosh/laravel-debug-eval
A rudimentary UI with backend for evaluating PHP remotely. For debugging purposes only.
1
33,136
padosoft/eval-harness
Laravel evaluation framework for RAG / LLM applications: golden datasets, twenty-two metrics including agent trajectories, repeated sampling with real statistics, per-row regression gates on content hashes, run briefings, cost budgets, JSON + Markdown reports, Artisan-driven CI gate.
7
9,671
mosaiqo/proofread
The only eval package native to the official Laravel AI stack. Evaluate agents, prompts, and MCP tools from Pest, CI, and production.
3,075
jkudish/pest-plugin-ai-benchmarks
A Pest plugin for comparative AI benchmarks, durable evidence, replay, and baselines.
1,515
padosoft/eval-harness-ai-bridge
Bridge between padosoft/eval-harness and the laravel/ai SDK: turn an AgentResponse into a scoreable trajectory, evaluate multi-turn conversations, and assert an agent against a golden dataset from Pest or PHPUnit.
4
offset-smoker/laravel-llm-evals
Evaluation harness for LLM features in Laravel: golden datasets, code-based graders, a required falsifiability control, CI exit codes, and a Claude driver built on the official Anthropic PHP SDK.
2