evals — independent software & tools

16 independent evals listings the big search engines bury.

• skval

  github · Free · by DCCA

  Score Claude Code skills with deterministic checks, safety gates, eval fixtures, and ship/revise/reject scorecards.

• brief-spec

  github · Free · by luanmorenommaciel

  Type-aware, evidence-backed delivery contract for AI coding harnesses. Different agents in. One predictable human handoff out.

• leadline

  github · Free · by Nazim22

  Evidence router & policy engine for coding agents — enforces source choice and proof-of-use through Claude Code hooks, including routes to MCP tools. Local-first; no LLM in the hook loop.

• ca-roads-mcp

  github · Free · by nicglazkov

  Live California road map, route planner, and AI assistant - plus an MCP server so any assistant gets CHP incidents, Caltrans closures, chain controls, cameras, and wildfires.

• modelpin

  github · Free · by samarthputhraya

  Dependabot for AI models — replay your app's scenarios across a model migration and catch real behavioral regressions before they ship. Low false-positive by design. CLI + GitHub Action. Open-core, Apache-2.0.

• cap-evolve

  github · Free · by skillberry-ai

  Optimize any AI agent’s skills, tools/MCP, and prompts against your own evals.

• elite-reasoning-mcp

  github · Free · by Snehgabani

  90+ tool Model Context Protocol server for AI coding agents: workflow flight recorder, quality-gated memory, release doctor, eval harnesses, and reasoning safety.

• crux

  github · Free · by use-crux

  Typed building blocks for everything around your LLM call.

• llmgauge

  github · Free · by WumboLabs

  Practical local LLM evaluation CLI for reproducible GGUF/llama.cpp testing on real hardware.

• eve

  npm · Free · by GitHub Actions

  Filesystem-first framework for durable backend AI agents that run anywhere.

• @mastra/core

  npm · Free · by GitHub Actions

  Mastra is a framework for building AI-powered applications and agents with a modern TypeScript stack.

• cam-cli

  github · Free · by eduardocaminha

  Local-first runtime for verifiable coding-agent loops.

• themoltnet

  github · Free · by getlarge

  Accountable authority for autonomous agents

• fieldwork

  github · Free · by ryanbaumann

  Fieldwork by Ryan Baumann: developer tools, Field Notes, working Labs, and agent evals.

• vigiles

  github · Free · by zernie

  Like Lighthouse for your agent harness - verify your CLAUDE.md/AGENTS.md, skills & hooks are real, then test and measure they actually work. Claude Code + Codex.

• evalite

  npm · Free · by mpocock

  Test your LLM-powered apps with a TypeScript-native, Vitest-based eval runner. No API key required.