Dev
GitHub repos gaining traction - what high-signal users are starring and what's climbing the board, captured daily and enriched from GitHub. Raw material for spotting new tech and patterns worth building on.
2,991
repos tracked
117
surfaced this week
232
created < 30d
Python
top language
19 repos
-
The first exam for code agents in the physical world: frozen robot tasks from many benchmark suites, each answered with a program and verified in a separate sandbox. Runs on Harbor.
-
Framework for evaluating and improving agents
-
Harbor task data backing tasksets in prime-envs
-
Writable-rootfs Singularity environment for harbor on FUSE-restricted HPC clusters (SWE-bench on PBS/SLURM, no Docker/Modal).
-
★ 0 neubig/harborHarbor is a framework for running agent evaluations and creating and using RL environments.
-
Writable-rootfs Singularity environment for harbor on FUSE-restricted HPC clusters (SWE-bench on PBS/SLURM, no Docker/Modal).
-
Terminal-Bench-Science: Evaluating AI agents on research workflows across scientific domains
-
★ 0 xeophon/harborHarbor is a framework for running agent evaluations and creating and using RL environments.
-
Measuring and evolving with the frontier of agent work
-
Harbor is a framework for running agent evaluations and creating and using RL environments.
-
A trajectory visualization demo using Harbor Task format and ATIF Agent Trajectory visualization.
-
Workspace manager for coding agents. Interactively solve and develop Harbor tasks.
-
Measuring and evolving with the frontier of agent work
-
Turn any repository into verifiable RL environments for coding agents - Harbor tasks you can train on, evaluate and share on the Hugging Face Hub
-
Pier is a Harbor fork built for DeepSWE, with stronger support for CLI agents in air-gapped (no-internet) tasks and more faithful, consistent agent trajectories
-
Realistic examples of building evals and optimizing agents with Harbor
-
Harbor dataset for Senior SWE-Bench (v2026.06)