Show HN: H5i-Python: Python SDK for Programmable Multi-Agent Orchestration

H5i-Python, a new Python SDK for the h5i orchestration engine, enables developers to define and execute multi-agent coding workflows across AI coding agents like Claude Code and Codex. The tool allows agents to work in sandboxed Git worktrees, with support for parallel implementation, mutual review, automated testing, and winner selection, aiming to create reproducible and auditable development processes.

Claude Code, Codex, and other coding agents have different strengths. However, naive multi-agent orchestration such as simply launching several agents in parallel or allowing them to exchange messages does not define a reproducible development process. A real workflow must specify: - who implements; - who reviews whom; - when an agent must revise its work; - which candidates are independently tested; - how the winner is selected; and - when the selected change is applied to the original branch. h5i-python is the Python SDK for the h5i https://github.com/h5i-dev/h5i orchestra engine. This SDK lets you define and execute multi-agent coding workflows across Claude Code, Codex, and other runtimes as ordinary Python programs. Each agent works inside its own sandboxed Git worktree, so it cannot overwrite the original checkout or another agent's work. Agent turns produce Git-backed artifacts that can be reviewed, revised, neutrally verified, compared, selected, and applied as one auditable workflow. Install the h5i https://github.com/h5i-dev/h5i engine: curl -fsSL https://raw.githubusercontent.com/h5i-dev/h5i/main/install.sh | sh cargo install --git https://github.com/h5i-dev/h5i h5i-core Install the Python SDK from GitHub: pip install h5i-orchestra pip install "git+https://github.com/h5i-dev/h5i-python.git" Create ensemble.py inside the Git repository the agents should modify. This workflow let Claude and Codex independently implement the same task, review and improve each other’s work, and then select the better result. python from h5i.orchestra import Conductor async def main task : async with Conductor repo=".", run="demo-task", launcher="resident" as c: claude = await c.hire "claude-agent", runtime="claude" codex = await c.hire "codex-agent", runtime="codex" Have both agents implement the task independently and in parallel claude work, codex work = await asyncio.gather claude.work task , codex.work task await c.freeze Seal the round, ensuring that neither agent influenced the other beforehand Have each agent review the other's work await asyncio.gather codex.review claude work , claude.review codex work Verify each submission in a fresh, neutral sandbox await c.verify claude work, "pytest", "--quiet" await c.verify codex work, "pytest", "--quiet" verdict = await c.judge Select the smallest diff among the submissions that pass all tests print "winner:", verdict.selected submission asyncio.run main "implement quicksort in python with unit test" Run it as a normal Python program: python ensemble.py With the default launcher="resident" , h5i automatically starts the agent sessions through tmux . For example, you can program: - ask Claude and Codex to implement the same task independently, have them review and improve each other's work, and select the smallest candidate that passes the tests; - let Claude Fable and Codex GPT-5.6 Sol iteratively refine a design, then hand the agreed design to Claude Opus for implementation; or - repeat a Fable-design/Sol-review loop ten times, ask Opus to implement the result, and invoke Sol to repair the implementation only when Fable rejects it. See examples/ /h5i-dev/h5i-python/blob/main/examples for complete scores, including: arena score.py /h5i-dev/h5i-python/blob/main/examples/arena score.py : independent arena ranking; ensemble score.py /h5i-dev/h5i-python/blob/main/examples/ensemble score.py : mutual-review ensembles; debate then build.py /h5i-dev/h5i-python/blob/main/examples/debate then build.py : architect-to-implementer pipelines; review escalation.py /h5i-dev/h5i-python/blob/main/examples/review escalation.py : conditional review escalation; judge panel score.py /h5i-dev/h5i-python/blob/main/examples/judge panel score.py : LLM judge panels; tournament.py /h5i-dev/h5i-python/blob/main/examples/tournament.py : tournament brackets; and custom control flow.py /h5i-dev/h5i-python/blob/main/examples/custom control flow.py : custom Python control flow. Apache-2.0