{"name":"Valkyrie","description":"Run scalable, reproducible evaluations for AI agents.","url":"https://docs.valkyrie.vals.ai/","version":"1.0.0","protocolVersion":"0.3","preferredTransport":"HTTP+JSON","supportedInterfaces":[{"url":"https://docs.valkyrie.vals.ai/","protocolBinding":"HTTP+JSON","protocolVersion":"0.3"}],"provider":{"url":"https://docs.valkyrie.vals.ai/","organization":"Valkyrie"},"documentationUrl":"https://docs.valkyrie.vals.ai/","capabilities":{"streaming":false,"pushNotifications":false},"defaultInputModes":["text/plain"],"defaultOutputModes":["text/plain"],"skills":[{"id":"valkyrie","name":"valkyrie","description":"Use when running agentic benchmarks at scale, managing agent deployments, monitoring benchmark runs, and retrieving evaluation results. Reach for this skill when you need to execute agents against benchmark datasets in isolated sandboxes, configure credentials and sandbox providers, or troubleshoot run failures.","tags":[],"url":"https://docs.valkyrie.vals.ai/.well-known/agent-skills/valkyrie/skill.md"}]}