{"name":"TunePlane","description":"TunePlane — one control plane for the whole LLM post-training lifecycle: environments, training, evaluation, serving and deployment, across any framework and any backend.","url":"https://docs.tuneplane.ai/","version":"1.0.0","protocolVersion":"0.3","preferredTransport":"HTTP+JSON","supportedInterfaces":[{"url":"https://docs.tuneplane.ai/","protocolBinding":"HTTP+JSON","protocolVersion":"0.3"}],"provider":{"url":"https://docs.tuneplane.ai/","organization":"TunePlane"},"documentationUrl":"https://docs.tuneplane.ai/","capabilities":{"streaming":false,"pushNotifications":false},"defaultInputModes":["text/plain"],"defaultOutputModes":["text/plain"],"skills":[{"id":"tune-plane","name":"TunePlane","description":"Use when submitting LLM post-training jobs (SFT, DPO, GRPO, reward modeling), running hyperparameter sweeps, evaluating models against benchmarks, deploying models for inference, managing datasets and volumes, or administering a TunePlane deployment. Agents should reach for this skill when users need to train models on shared GPU clusters, track experiments, score outputs, or serve models.","tags":[],"url":"https://docs.tuneplane.ai/.well-known/agent-skills/tune-plane/skill.md"}]}