From 65ced6b2f0dd25b98bc2792702902e64820ec722 Mon Sep 17 00:00:00 2001 From: benediktstroebl Date: Fri, 27 Mar 2026 15:33:19 -0700 Subject: [PATCH 1/2] Add Prime RL optimization example --- README.md | 1 + harbor_cookbook/prime_rl/README.md | 23 +++++++++++++++++++++++ 2 files changed, 24 insertions(+) create mode 100644 harbor_cookbook/prime_rl/README.md diff --git a/README.md b/README.md index 458c76c..ae9d681 100644 --- a/README.md +++ b/README.md @@ -39,4 +39,5 @@ harbor run -p harbor_cookbook/recipes/ -a claude-code -m anthropic/claude- |:--|:--| | [gepa](harbor_cookbook/gepa/) | Agent harness optimization for MedAgentBench using Harbor+GEPA. | | [tinker‑rl](harbor_cookbook/tinker_rl/) | RL training on Harbor tasks using the Tinker SDK. | +| [prime‑rl](harbor_cookbook/prime_rl/) | RL training on Harbor tasks using Prime RL and Verifiers. | | [sky‑rl](harbor_cookbook/sky_rl/) | RL training on Harbor tasks using SkyRL. | diff --git a/harbor_cookbook/prime_rl/README.md b/harbor_cookbook/prime_rl/README.md new file mode 100644 index 0000000..be41e35 --- /dev/null +++ b/harbor_cookbook/prime_rl/README.md @@ -0,0 +1,23 @@ +# prime_rl + +RL training on Harbor tasks using [Prime RL](https://docs.primeintellect.ai/prime-rl). + +This directory is a pointer. The full environment integration lives in the verifiers repo: + +**[verifiers/environments/opencode_harbor](https://github.com/PrimeIntellect-ai/verifiers/tree/main/environments/opencode_harbor)** + +## Quick start + +Set up your workspace: + +```bash +prime lab setup --prime-rl +``` + +Configure your environment TOML to use the `opencode_harbor` environment, then launch training: + +```bash +prime lab start +``` + +See the [opencode_harbor README](https://github.com/PrimeIntellect-ai/verifiers/tree/main/environments/opencode_harbor) for full setup instructions. From 9fb872689be9833891e719b5235935c06d839561 Mon Sep 17 00:00:00 2001 From: benediktstroebl Date: Fri, 27 Mar 2026 15:37:50 -0700 Subject: [PATCH 2/2] Simplify Prime RL README --- harbor_cookbook/prime_rl/README.md | 20 +------------------- 1 file changed, 1 insertion(+), 19 deletions(-) diff --git a/harbor_cookbook/prime_rl/README.md b/harbor_cookbook/prime_rl/README.md index be41e35..8c11dd0 100644 --- a/harbor_cookbook/prime_rl/README.md +++ b/harbor_cookbook/prime_rl/README.md @@ -2,22 +2,4 @@ RL training on Harbor tasks using [Prime RL](https://docs.primeintellect.ai/prime-rl). -This directory is a pointer. The full environment integration lives in the verifiers repo: - -**[verifiers/environments/opencode_harbor](https://github.com/PrimeIntellect-ai/verifiers/tree/main/environments/opencode_harbor)** - -## Quick start - -Set up your workspace: - -```bash -prime lab setup --prime-rl -``` - -Configure your environment TOML to use the `opencode_harbor` environment, then launch training: - -```bash -prime lab start -``` - -See the [opencode_harbor README](https://github.com/PrimeIntellect-ai/verifiers/tree/main/environments/opencode_harbor) for full setup instructions. +This directory is a pointer. See the [opencode_harbor README](https://github.com/PrimeIntellect-ai/verifiers/tree/main/environments/opencode_harbor) for the full example and the setup instructions.