Updated today
by yatuk
multi-model-orchestra
Route coding work across Claude, GPT (Codex), Kimi and DeepSeek: one runner with hard status gates, worktree-only writes, a blind council and a benchmark.
29 GitHub stars (29)
+29 in 11 days
- 6 skills (6 skills found in its SKILL.md files)
- 2 forks (2 forks)
- Today last update (Last updated today)
Install
Skills CLI
npx skills add yatuk/multi-model-orchestra
Asks which skills and agents to install. Just one: npx skills add yatuk/multi-model-orchestra --skill chatgpt-lane — add -g for every project.
Claude Code
git clone https://github.com/yatuk/multi-model-orchestracp -r "multi-model-orchestra/skills/chatgpt-lane" ~/.claude/skills/
Copies the chatgpt-lane skill into ~/.claude/skills/ for every project (use .claude/skills/ for one project). Repeat for the other skills listed below.
Plugin
/plugin marketplace add yatuk/multi-model-orchestra
Run inside Claude Code if the repository ships a plugin marketplace, then pick plugins from /plugin.
Skills can run code on your machine — read the SKILL.md first.
What it does
In the author's own words, from each skill's SKILL.md.
-
chatgpt-lane
Run a job on OpenAI GPT models through the Codex CLI (gpt-6.1-sol for execution, gpt-6-astra for second opinions, gpt-6-luna for bulk). Use when the user asks for GPT, Codex or ChatGPT, or when the orchestra routing table picks the GPT side. Needs the orchestra skill's scripts.
skills/chatgpt-lane/ on GitHub (opens in a new tab)SKILL.md -
council
Ask the same question to three models from different families (GPT, Kimi, DeepSeek by default), read-only and in parallel, then judge the answers blind. Use for design decisions, risky trade-offs, root-cause hypotheses and reviews where one model's view is not enough. Needs the orchestra skill's scripts.
skills/council/ on GitHub (opens in a new tab)SKILL.md -
deepseek-lane
Run a job on DeepSeek models (deepseek-v4-pro, deepseek-flash) through the Kimi Code CLI with a DeepSeek provider configured: an independent opinion from another model family, cheap execution, or bulk labeling. Use when the user asks for DeepSeek or the orchestra routing table picks it. Needs the orchestra skill's scripts.
skills/deepseek-lane/ on GitHub (opens in a new tab)SKILL.md -
kimi-lane
Run a job on Kimi models through the Kimi Code CLI (kimi -p): cheap execution of easy, testable work in its own git worktree, or a quick extra opinion. Use when the user asks for Kimi, when other lanes hit their limits, or when the orchestra routing table picks Kimi. Needs the orchestra skill's scripts.
skills/kimi-lane/ on GitHub (opens in a new tab)SKILL.md -
orchestra
Routing policy for a coding agent that can hand work to other models: Claude tiers (Opus, Sonnet, Haiku), OpenAI GPT through the Codex CLI, Kimi and DeepSeek through the Kimi Code CLI. Decides who does a job, how to brief it, how to run it safely (one runner, hard status gates, worktree per writing lane) and how to check the answer before using it. Load before spawning any sub-agent or external m…
skills/orchestra/ on GitHub (opens in a new tab)SKILL.md -
otomode
Unattended work mode for when the user is away. `otomode 6h` asks everything up front in one round (tasks, how to check each, permissions), writes OTOMODE.md, turns on a hook that denies every permission prompt and question and logs it to QUESTIONS.md, then works task by task through subagents (Claude, DeepSeek), verifies each with its own check and commits each separately, pauses at 85% usage an…
skills/otomode/ on GitHub (opens in a new tab)SKILL.md
No skill matches that filter.
Review before you install
A skill is a set of instructions and scripts your agent will follow, and it can run code on your machine. Read the SKILL.md and any scripts first — this listing is not a security review.