Instructions file CodexOpenCode
Instructions for cubeplexai/cubepi, covering cubepi, commands (always via uv), architecture, conventions & gotchas and development workflow.
Instructions file CodexOpenCode
Instructions for cubeplexai/cubepi, covering cubepi, commands (always via uv), architecture, conventions & gotchas and development workflow.
Instructions file
Instructions for cubeplexai/cubepi, a project described as: CubePi — An async-first Python agent kernel inspired by Pi, built for coding agents, tools, and continuous sessions.
Skill Claude CodeCodex
Use when debugging a cubepi/cubebox agent run — a missing final reply, a tool that misbehaved, a 4xx from the model, wrong token/cache numbers, or "why did the agent do that?". Reads the JSONL traces cubepi writes per run via the cubepi trace CLI (ls / view / follow / stats) to inspect the span tree, errors, tool…
Skill Claude CodeCodex
Use when building, extending, or debugging agents with the CubePi framework. Covers the Agent API, providers, tools, middleware, checkpointing, MCP integration, and HITL. References the cubepi-trace skill for run debugging.
Agent
Create and run your first CubePi agent — choose a provider, define a system prompt, and send a message.
Agent
Branch a conversation at a completed-run boundary — persistent forks for new threads, ephemeral one-shot probes, and the HITL binding rules that keep them safe.
Agent
Build multi-turn conversational agents with CubePi's stateful agent loop and message history.
Agent
How CubePi maximises prompt cache hit rates — append-only message storage, automatic cache breakpoints, and how to avoid breaking the cache.
Agent
Stream events and tokens from CubePi agents in real time using subscribers and MessageStream.
Agent
Register tools, execute them in parallel or sequentially, and handle results with Pydantic validation in CubePi.