shepherd

ai-tools / GitHub

SCORE

Shepherd is an AI framework focused on runtime supervision and meta-optimization of agents. It leverages Monte Carlo Tree Search reinforcement learning (MCTS-RL) techniques to guide and improve agent behavior during execution. The project supports meta-agent architectures for tree-based RL workflows.

VISIT PRODUCT →

Sources (1)

GitHub

https://github.com/shepherd-agents/shepherd

674 STARS

Score Breakdown

Traction

raw 714.00 · weight 35%

33.3pts

Novelty

0.7 days old · weight 20%

20.0pts

Source diversity

1 source · weight 10%

3.3pts

AI quality

raw 68.00 · weight 35%

23.8pts

Technically differentiated MCTS-RL agent supervision framework with decent GitHub traction (674 stars, 40 forks) but niche appeal and limited community engagement.

Final Score80/100