RepoAlerts

garrytan/g-harness-evals

Resource

Can I trust it?

Worth a closer look before you use it.

  • Warnings Check

    Very new: it was created 12 days ago. Wait a little before installing it.

  • Stars from real accounts No data

    We haven't checked its stars for fakes yet.

  • Active OK

    Last change yesterday.

  • Licence OK

    MIT: you can use it for free, also in a commercial product. Keep the licence notice.

Quick checks from public data. They help, but they are no guarantee.

What it is

A benchmark comparing AI agent harnesses (the system prompt, tools and memory wrapped around a model) on the same frontier model and tasks, measuring pass rate, cost per solved task and time. Reports coding results on 52 Terminal-Bench 2.1 tasks and multi-user tasks, with…

Who it's for

Developers choosing an AI coding agent harness and comparing cost per solved task

How to use it

Ask your AI to set it up

Paste this into Claude, ChatGPT or your coding assistant.

Help me install and set up garrytan/g-harness-evals in my project. Follow its README: https://github.com/garrytan/g-harness-evals. Explain each step and check it is safe before running anything.

Read the README

Star curve

157From 0 to 157 stars between 5 Nov 2026 and 18 Nov 2026.

Get alerts for repos like this

We'll email you when something in AI agents starts rising. Free, once a week, only what matches.

What arrives: one email on Mondays with the repos that match, never more. See last week's. Privacy

For maintainers

Add this badge to your README. It shows the repo's place on the radar, updates every hour and links back to this page.

[![garrytan/g-harness-evals on RepoAlerts](https://repoalerts.com/badge/garrytan/g-harness-evals.svg)](https://repoalerts.com/repo/garrytan/g-harness-evals?ref=badge)
Share on X

  • google-research/harness2

    A Python framework for recursive agent harnessing, which refines the prompts, tools, memory, skills and orchestration around a fixed model. It runs a propose-probe-compose loop per task and keeps a pe

    AgentAI agentsPython · 24,309 stars · +896 this week · 42 days old · HN 81 pts

  • pzs19/HarnessPro

    A framework that recursively improves the harness of an AI agent (its instructions, tools and runtime policies) by diagnosing execution traces and selecting candidate changes with synthesized tests. O

    ResourceAI agentsPython · 173 stars · +13 this week · 55 days old

  • rsi-gear/gear

    A TypeScript framework that optimizes AI agent harnesses (instructions, tools, workflows and model settings) on a benchmark of tasks. A Meta agent analyzes execution trajectories and evaluation feedba

    AI skillAI agentsTypeScript · 476 stars · +36 this week · 92 days old

  • YuhuaJiang2002/HarnessMM

    Research implementation of HarnessMM, which reuses decisions an agent already made: a slow model call records future tool actions in a decision graph, and guarded checks on tool results decide whether

    AgentAI agentsPython · 194 stars · +36 this week · 23 days old

  • zgcagi/Harness-RSI

    A survey and companion reading list on how execution experience becomes persistent harness state for a fixed base model. It covers update targets, evolutionary search, theory, evaluation, applications

    ResourceAI agentsTeX · 451 stars · +4 this week · 48 days old

  • IsuruMaduranga/harness-engineering-101

    An open book series in Markdown that explains how AI agents work by building a roughly 300-line toy agent harness in plain Python. It covers the message array, tool calls, the agent loop, context mana

    ResourceAI agentsPython · 657 stars · +5 this week · 72 days old · HN 132 pts