LLM backend abstraction. Today the only implementation is ClaudeCodeBackend (server/agent/backend/claude-code.ts), which spawns the `claude` CLI as a subprocess. The interface exists so future backends (OpenAI, Ollama native, Gemini, etc.) can plug in here without the orchestrato
// Source: https://github.com/receptron/mulmoclaude | License: MIT | Copyright 2026 Satoshi Nakajima and the members of Receptron
// Provided via justclaw.it.com; keep this notice with the code.
// LLM backend abstraction. Today the only implementation is
// ClaudeCodeBackend (server/agent/backend/claude-code.ts), which spawns
// the `claude` CLI as a subprocess. The interface exists so future
// backends (OpenAI, Ollama native, Gemini, etc.) can plug in here
// without the orchestrator in server/agent/index.ts knowing which one
// is active.
//
// See plans/done/refactor-llm-backend-abstraction.md for the broader plan.
import type { Attachment } from "@mulmobridge/protocol";
import type { Role } from "../../../src/config/roles.js";
import type { ChatModel, EffortLevel } from "../../../sr
... (truncated -- full source via MCP)
See the full source, get the GitHub permalink, and search for more like it.
Get a free API key