Skip to main content
Pre-release

This release is a pre-release and may not be stable for production use.

Ordin

Ordin 是模型中立、单进程生命周期的 Python Agent 编排库。平台先持久化 UserMessage,再把 ExecutionRequest 交给一个 Ordin 对象;根主控和开发者 Agent 都使用同一个公共 Agent 类型和同一个 AgentLifecycle。未设置 custom logic 时,它们共享默认的 Context/模型/Tool/Artifact 多轮循环。

根 Agent 可以通过受控 Tool 创建 Task DAG、选择已注册 Agent 与模型、并发执行子 Agent、等待首个或 全部结果,并在检查和取消旧执行后原子修订 Plan。Provider、Store、ObjectStore 和注册表由 Ordin 持有,不进入 Agent 定义。

安装

pip install ordin

最小示例

from ordin import Agent, Ordin
from ordin.common import Identity
from ordin.llm import LLMCallPipeline, LLMProfile
from ordin.observability import EventBroker
from ordin.testing import InMemoryObjectStore, InMemoryOrdinBackend

backend = InMemoryOrdinBackend()
events = EventBroker(backend)
model = LLMCallPipeline(
    provider=my_llm_provider,
    recorder=backend,
    event_sink=events,
).bind(LLMProfile(id="private-model"))

ordin = Ordin(
    root_agent=Agent[None](
        name="root",
        instructions="Answer directly or create bounded child Agents when useful.",
        model=model.id,
    ),
    models=(model,),
    backend=backend,
    object_store=InMemoryObjectStore(),
    events=events,
)

ordin.register(
    Agent[None](
        name="research",
        description="Produce bounded research evidence.",
        instructions="Complete only the current Task.",
        model=model.id,
    )
)

request = await backend.prepare_execution(
    "分析这个问题",
    identity=Identity(user_id="u-1"),
)
result = await ordin.execute(request, context=None)

完整离线示例见 examples/direct.pyexamples/mission.pyexamples/custom_agent.py;运行中取消与原子修订见 examples/plan_revision.py

文档

开发校验

uv sync
uv run python scripts/docs.py --check
uv run pytest tests/test_docs.py
uv run ruff format .
uv run ruff check .
uv run ty check
uv run pytest

Release files for ordin 0.1.1.dev1635051000000006001

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for ordin 0.1.1.dev1635051000000006001
File Size Uploaded
ordin-0.1.1.dev1635051000000006001.tar.gz 217.4 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for ordin 0.1.1.dev1635051000000006001
File Interpreter ABI Platform
ordin-0.1.1.dev1635051000000006001-py3-none-any.whl Python 3 none any Details

Total release size: 333.0 kB

Release files / ordin-0.1.1.dev1635051000000006001.tar.gz

Download URL ordin-0.1.1.dev1635051000000006001.tar.gz
Size 217.4 kB
Tags Source
SHA-256 checksum
How to use checksums
607905a173cf8416a96641e453e1eb56475bbf66259b92e61c9a4b47b9f2722a
BLAKE2b-256 checksum
How to use checksums
c1aa5d4796519f94e49af0e4249e14fc8c9c048f783cf106146c92a3a5e8252f
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via uv/0.12.5 {"installer":{"name":"uv","version":"0.12.5","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Aug 27, 2026.

Transparency log

Release files / ordin-0.1.1.dev1635051000000006001-py3-none-any.whl

Download URL ordin-0.1.1.dev1635051000000006001-py3-none-any.whl
Size 115.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
51e9351fb844788ab44754f7c084558772fb66a115d34a297f13e1e44acb8bee
BLAKE2b-256 checksum
How to use checksums
cacd6feeb16c561bb658bbdf665316ecc9d86560115baf91049f64b7c259447d
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via uv/0.12.5 {"installer":{"name":"uv","version":"0.12.5","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Aug 27, 2026.

Transparency log
Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page