mirror of
https://github.com/sipeed/picoclaw.git
synced 2026-06-12 18:08:54 +00:00
xiaoen
8034ee7be1
Two estimation bugs fixed: 1. Media tokens were added to the chars accumulator before the chars*2/5 conversion, resulting in 256*2/5=102 tokens per item instead of 256. Fix: add media tokens directly to the final token count, bypassing the character-based heuristic. 2. estimateMessageTokens counted both tc.Name and tc.Function.Name for tool calls, but providers only send one (OpenAI-compat uses function.name, Anthropic uses tc.Name). Fix: count tc.Function.Name when Function is present, fall back to tc.Name only otherwise. Also fix i18n hint text: "auto-detect" was misleading — the backend uses a 4x max_tokens heuristic, not actual model detection.
Picoclaw Web
This directory contains the standalone web service for picoclaw.
It provides a complete unified web interface, acting as a dashboard, configuration center, and interactive console (channel client) for the core picoclaw engine.
Architecture
The service is structured as a monorepo containing both the backend and frontend code to ensure high cohesion and simplify deployment.
backend/: The Go-based web server. It provides RESTful APIs, manages WebSocket connections for chat, and handles the lifecycle of thepicoclawprocess. It eventually embeds the compiled frontend assets into a single executable.frontend/: The Vite + React + TanStack Router single-page application (SPA). It provides the interactive user interface.
Getting Started
Prerequisites
- Go 1.25+
- Node.js 20+ with pnpm
Development
Run both the frontend dev server and the Go backend simultaneously:
make dev
Or run them separately:
make dev-frontend # Vite dev server
make dev-backend # Go backend
Build
Build the frontend and embed it into a single Go binary:
make build
The output binary is backend/picoclaw-web.
Other Commands
make test # Run backend tests and frontend lint
make lint # Run go vet and prettier/eslint
make clean # Remove all build artifacts