A Claude Code plugin of Harbor skills for building and evaluating LLM agents on long-horizon terminal tasks. It covers creating benchmark adapters, writing tasks and verifiers, and publishing tasks or datasets to the Harbor registry.
@zli12321Scaffold a new Harbor benchmark adapter by running `harbor adapter init` and then guide implementation using the Adapters Agent…
@zli12321Create a new Harbor task for evaluating agents. Use when the user wants to
@zli12321Publish a Harbor task or dataset to the registry. Use when the user wants to upload, publish, or share tasks or…
@zli12321Write Harbor task verifiers using Reward Kit. Use when creating or editing a
@zli12321Create or reuse Hugging Face dataset PRs for `harborframework/parity-experiments` and upload Harbor parity/oracle result folders…
@zli12321Is this you? Claim your profilePut a plugin at the top of Plugins, from $9
