AI Runtime & Agent Substrate
agent-runtime
Lower-level agent execution — message reducer, tool invocation, streaming state machine. The kernel @nebutra/agents wraps.
Agent runclaude-sonnet-5
Ready- User promptFind recent signups in EU and email the team
- Model turnplanning · 412 tokens
- tool_callsearch_users({ region: 'eu', since: '7d' })
- tool_result→ 23 rows · 184 ms
- tool_callsend_email({ to: 'team@', body: ... })
- tool_result→ ok · message_id=msg_8f2a
- Final answerSent digest to team. 23 EU signups.
Steps
7
Tokens
1,247
Latency
2.3s
Tool calls
2
More in nebutra.com
- 3d-pipeline3D asset generation + render — text-to-mesh + USD export + Three.js preview, with GPU-pool routing on the backend.
- agentsMulti-step agent runtime — Vercel AI SDK foundation, tool registry, streaming UI, MCP integration. The skeleton for every Sailor agentic feature.
- ai-primitivesShared helpers used across all AI packages — token counting, prompt templating, stream-parser, retry-with-backoff. No provider dependency.
- ai-providersProvider metadata — pricing per million tokens, context windows, modalities. Drives model selection and cost estimates.
- atelier-canvasInfinite design canvas — multiplayer cursors, layer panel, frame export. The drawing surface every atelier play renders into.
- audio-pipelineAudio processing chain — voice clone, denoise, transcribe, summarize. Streams through one async iterator per stage.