Wisp Router — AI skill for Claude Code
A model router for your Agent Harness (e.g Claude Code).
How to install Wisp Router
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open EstarinAzx/Wisp-Router and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Wisp Router does
A model router for your Agent Harness (e.g Claude Code). Bring your own backend and run your favourite models.
Alternatives in AI
- Bot On Anything — A large model-based chatbot builder that can quickly integrate AI models (including ChatGPT, Claude, Gemini) i 4.2k ★
- Manifest — Real-time cost observability for OpenClaw agents — track tokens, costs, messages, and model usage 4.1k ★
- Pilotfish — Multi-model orchestration layer for Claude Code — the frontier model plans, cheaper models execute, verificati 688 ★
README
BYOK model router — run the model access you already pay for, everywhere you code.
What is Wisp
Wisp routes your own model backends — your **ChatGPT (Codex) subscription**, your **Claude.ai subscription** (both via OAuth, no API key), or any OpenAI-compatible API key (OpenAI, Groq, Mistral, OpenRouter, Ollama, and more) — into the tools you already code with:
- VS Code — providers show up as native models in Copilot's Chat view, Agent mode, and the
Ctrl+Ipicker, with streaming, tool calling, vision, and live per-model context windows. - Claude Code — the Bridge exposes the same providers as a local endpoint speaking two dialects (OpenAI
/v1/chat/completions+ Anthropic `/v1/messa
Related Skills
Pragma
Build portable agent teams across models and harnesses, with shared memory, skills, tools, and workflows — the
V Models
Refresh the Compound V tier→model map — discover the concrete models each backend (claude, codex, antigravity,
Claude Code Model Router
A lightweight API gateway that lets you switch to third-party AI models when using Claude Code.
Harness Arena
Blind, LMSYS-style arena for comparing AI coding agent harnesses (Claude Code, Codex CLI, OpenClaw, Hermes, On
Persona Portability Benchmark
One persona, one frozen memory, N models: how much of an agent's character survives a model swap? Harness + bl
Harness Arena Fe
Blind, LMSYS-style arena for comparing AI coding agent harnesses (Claude Code, Codex CLI, OpenClaw, Hermes, On
Related Agents
Model Scout
Reviews AI model candidates and existing catalog entries for the model library at ~/.claude/router/models/ (ca
DB Seeder
Writes and maintains backend/seed.py, the Faker-based script that fills a development database with realistic
Parity Porter
Port one module of a PyTorch model to ttnn and prove it matches its CPU golden. Use for a single component in