Enables external agents to list T3 environments and projects, start coding threads, monitor progress, send follow-up instructions, and interrupt running turns via MCP.
Enables any agent to spawn and manage T3 Code agent threads over HTTP MCP — starting research or coding agents, sending follow-ups, waiting for results, and answering approval prompts. It pairs once as a T3 device, exposes 15 tools, and keeps operations retry-safe.
Enables MCP clients to create, read, message, wait for, interrupt, and settle persistent peer threads through an existing T3 Code server, keeping them visible across its clients. It supports project-scoped coordination with source attribution and retry-safe commands.
MCP server for interacting with a running T3 Code instance. Enables viewing agent threads, sending messages, and approving permission requests from Claude Code or voice-controlled models.
Enables Claude Code to delegate coding implementation and testing to a local pi agent via MCP tools, including dispatching task books, monitoring status, steering or aborting runs mid-execution, and retrieving results and transcripts.