Type
org
Skills
4
Claimed
No
Verified
No
Review boundary
Author information helps you judge provenance; different skills from the same author may still have different source states, so review each one before install.
Published skills
Inspect each skill on the cards before install.
Repository: harbor
Author: harbor-framework · Source status: Clear source
Harbor is a framework for running agent evaluations and creating and using RL environments.
Score basis：Clear source · Low risk signals · Universal
Repository: terminal-bench-science
Author: harbor-framework · Source status: Clear source
Terminal-Bench-Science: Evaluating AI Agents on Complex Real-World Scientific Workflows in the Terminal Topics: agentic-ai, ai-for-science, ai4science.
Score basis：Clear source · Low risk signals · Universal
Repository: harbor-cookbook
Author: harbor-framework · Source status: Clear source
Realistic examples of building evals and optimizing agents with Harbor
Score basis：Clear source · Low risk signals · Universal
