Skip to main content

agent-operations

中文 | English

Open-source Agent Operations (AgentOps / LLMOps) toolkit — trace runs, track latency / tokens / cost, and debug tool failures.

Lightweight and self-hostable. Default storage is one SQLite file. No ClickHouse required.

项 值
Python ≥ 3.11
Install pip install agent-operations(或源码 pip install -e .)
Default UI http://127.0.0.1:8787
Default DB ./data/agentops.db

什么人该用 / 不该用

一句话:这是最小可自托管的 Agent 运行账本,不是完整的 LLMOps 云平台。

更适合你,如果… 更该选别的(如 Langfuse / 云厂商控制台),如果…
已有 agent 代码,只想记 run / steps / 延迟 / token / 费用 / 失败 要漂亮的团队 UI、Prompt 实验、人人点选的分析台
想本机或内网跑,不想注册 SaaS、不想养 ClickHouse 要开箱即用、教程最多、出问题好搜
pip 一个包就能塞进现有循环 要托管采集、多区域、厂商级账单对账
单机 SQLite 就够;长大再迁 Postgres 一上来就要多租户产品级登录与权限中心

小白路径建议:本机 pip install agent-operations → agentops-smoke(应打印 smoke passed)→ examples/quickstart.py → agentops-http 看页面;确认有用再考虑云上部署。

系统要求(Windows / macOS / Linux / iOS)

环境 能否顺利用 说明
Windows 可以 PowerShell / cmd;推荐 Python 3.11+ 与 uv
macOS 可以 Terminal;同上
Linux 可以 发行版自带或官方 Python 3.11+
iPhone / iPad (iOS) 不作为安装端 不能在手机里 pip install 跑服务;手机浏览器可打开你已部署好的调试页;App 只能当 HTTP 客户端上报 run

核心依赖是 Python 3.11+,三桌面系统等价。云上部署后,任意能发 HTTPS 的客户端(含 iOS App)都能用,与本机是否装过本包装无关。

部署到云给其它服务用

可以。其它云上的 agent 不用把库装进对方机器里;只要能 HTTP 访问你的 agentops-http,把 run POST 过来即可。

[云 A 上的 agent] --HTTPS POST /api/runs--> [你部署的 agentops-http + DB]
[云 B 上的 worker] --同上-------------------┘

最小云上步骤(任意 VPS / 容器均可):

pip install "agent-operations[postgres]"   # 多机或托管库时建议 Postgres;单机也可用 SQLite 盘
$env:AGENTOPS_API_KEYS = '{"svc-a":{"projects":["demo"]},"admin":{"projects":["*"]}}'
# 多实例共享库时:
# $env:AGENTOPS_DATABASE_URL = "postgresql://user:pass@host:5432/agentops"
agentops-http --host 0.0.0.0 --port 8787 --db $env:AGENTOPS_DATABASE_URL

前面加 Nginx / Caddy 做 HTTPS,对外只暴露 443。调用方示例:

curl -s https://agentops.example.com/api/runs `
  -H "X-API-Key: svc-a" -H "Content-Type: application/json" `
  -d '{"id":"run-cloud-1","project":"demo","agent":"assistant","environment":"prd","name":"job","status":"ok","started_at":"2026-10-25T04:00:00+00:00","ended_at":"2026-10-25T04:00:01+00:00"}'

云上注意:

点 说明
鉴权 非本机绑定必须配置 API key;密钥当密码保管
库 SQLite = 单实例 + 持久盘;多机写同一数据用 Postgres
职责 只记账本,不替你调 OpenAI;各服务仍直连自己的 LLM
安全 不要把 --host 0.0.0.0 裸奔在公网无 HTTPS / 无密钥

5 分钟上手

1. 安装并冒烟

python -m pip install agent-operations
agentops-smoke
# 源码树也可用:python scripts/smoke.py

看到 smoke passed 即本机链路正常(Win / macOS / Linux 同一命令)。

从源码开发(推荐 uv):

git clone https://github.com/WenSongWang/agent-operations.git
cd agent-operations
uv venv .venv
.\.venv\Scripts\Activate.ps1
uv pip install -e ".[agentic-loop,openai,langgraph]"
python -m unittest discover -s tests -v

没有 uv 时:python -m venv .venv,激活后用 python -m pip install -e ".[agentic-loop,openai,langgraph]"。

2. 用 Python 写入一条 run

默认 SQLite 文件;Postgres 时把路径换成 DSN(需 pip install "agent-operations[postgres]"):

from agentops import connect, ingest_run

conn = connect("./data/agentops.db")
# conn = connect("postgresql://user:pass@localhost:5432/agentops")
ingest_run(conn, {
    "id": "run-001",
    "project": "demo",
    "agent": "assistant",
    "environment": "dev",          # dev | test | uat | prd
    "name": "answer-question",
    "status": "ok",                # running | ok | error | timeout | cancelled
    "started_at": "2026-10-24T10:00:00+00:00",
    "ended_at": "2026-10-24T10:00:02+00:00",
    "provider": "openai",
    "model": "gpt-4o-mini",
    "input_tokens": 100,
    "output_tokens": 50,
    # cost_usd 可省略:会按内置单价估算
    "steps": [
        {
            "seq": 1,
            "kind": "llm",         # llm | tool | agent | human
            "name": "gpt-4o-mini",
            "status": "ok",
            "started_at": "2026-10-24T10:00:00+00:00",
            "ended_at": "2026-10-24T10:00:02+00:00",
            "input_tokens": 100,
            "output_tokens": 50,
        }
    ],
})
conn.close()

或直接跑示例:

python examples/quickstart.py
python examples/openai_wrap.py

2b. 更省事:用 TracedRun(推荐)

不用手写 running / 终态时间戳;OpenAI 响应可直接抽 token:

from agentops import connect, TracedRun

conn = connect("./data/agentops.db")
with TracedRun(
    conn,
    project="demo",
    agent="assistant",
    environment="dev",
    name="chat",
    task_input="Say hi",
) as tr:
    # resp = client.chat.completions.create(...)
    # tr.record_openai(resp, output_text=resp.choices[0].message.content)
    tr.finish(task_output="hi")
conn.close()

异常退出时会自动记成 status=error。假响应示意:examples/openai_wrap.py。

真实调用(需 OPENAI_API_KEY + pip install openai):

$env:OPENAI_API_KEY = "sk-..."
# 可选:$env:OPENAI_BASE_URL = "https://api.openai.com/v1"
# 可选:$env:OPENAI_MODEL = "gpt-4o-mini"
python examples/openai_live.py

2c. 接 Agentic-Loop

pip install "agent-operations[agentic-loop]"

把 AgentOpsJournal 传给 run_loop(..., journal=...)(与官方 JSONL journal 同接口):

from agentops import connect
from agentops.adapters import AgentOpsJournal

conn = connect("./data/agentops.db")
journal = AgentOpsJournal(
    conn,
    project="demo",
    agent="assistant",
    environment="dev",
    runs_dir="./.agentic-loop/runs",  # 可选:同时写 agentic-loop JSONL
)
# terminal = run_loop(messages, tools=tools, llm=llm, journal=journal)

可运行示意:python examples/agentic_loop_wrap.py。

2d. 接 LangGraph

from agentops import connect
from agentops.adapters.langgraph import traced_invoke

conn = connect("./data/agentops.db")
result = traced_invoke(
    app,  # any object with .invoke / use traced_ainvoke for .ainvoke
    {"messages": [...]},
    conn=conn,
    project="demo",
    agent="support-bot",
    environment="dev",
    name="support",
    config={"configurable": {"thread_id": "1"}},
)

示意(无 LLM):python examples/langgraph_wrap.py。

3. 启动本机服务与页面

agentops-http
# 等价:python -m agentops.http
# Postgres(需 agent-operations[postgres]):
# agentops-http --db "postgresql://user:pass@localhost:5432/agentops"
# 或 $env:AGENTOPS_DATABASE_URL = "postgresql://..."

浏览器打开 http://127.0.0.1:8787

4. 用 HTTP / curl 上报

curl -s http://127.0.0.1:8787/api/runs -Method POST -ContentType "application/json" -Body @'
{
  "id": "run-002",
  "project": "demo",
  "agent": "assistant",
  "environment": "dev",
  "name": "via-http",
  "status": "ok",
  "started_at": "2026-10-24T11:00:00+00:00",
  "ended_at": "2026-10-24T11:00:01+00:00",
  "latency_ms": 1000,
  "input_tokens": 10,
  "output_tokens": 5
}
'@

查询:

curl -s "http://127.0.0.1:8787/api/runs?project=demo"
curl -s "http://127.0.0.1:8787/api/runs/summary?project=demo"
curl -s "http://127.0.0.1:8787/api/failures?project=demo"
curl -s "http://127.0.0.1:8787/api/runs/run-002"

API 一览

方法 路径 说明
GET / 只读调试页
POST /api/runs 写入 / 终态更新一条 run
GET /api/runs 列表(可按 project/agent/environment/status/时间筛选)
GET /api/runs/summary 次数、成功率、token/费用、p50/p95/p99
GET /api/runs/{id} 单条详情(含 steps)
GET /api/failures 工具失败归并

必填字段(run)

id, project, agent, environment, name, status, started_at

终态(ok / error / timeout / cancelled)还要 ended_at。
status=running 时可省略 ended_at;同一 id 再报终态会覆盖更新。

常用可选字段

字段 说明
latency_ms / ttft_ms 耗时;不传则尽量从时间戳 / steps 推算
input_tokens / output_tokens / cached_tokens / reasoning_tokens token;未上报保持空,不是 0
cost_usd 费用;不传则按 provider+model 估算
provider / model 厂商与模型
steps 步骤列表(排错用)
quality_score 0–1
thumbs up / down
prompt_version / template_version 提示词 / 模板版本
tags JSON,如 {"end_user_id":"..."}
extra 备用 JSON

步骤必填:seq, kind, name, status。

可选能力

脱敏

$env:AGENTOPS_REDACT = "hash"   # off | hash | drop

作用于 task_input / task_output、步骤 input / output / traceback、tags.end_user_id。

价目表

未传 cost_usd 时用内置单价估算。自定义 JSON(键为 provider/model):

$env:AGENTOPS_PRICING_PATH = ".\pricing.json"

告警(适合 cron)

$env:AGENTOPS_ALERT_WEBHOOK = "https://example.com/hook"
$env:AGENTOPS_ALERT_FAILURE_RATE = "0.2"
$env:AGENTOPS_ALERT_P95_MS = "3000"
$env:AGENTOPS_ALERT_COST_USD = "10"
# 可选:同时发邮件(stdlib SMTP)
$env:AGENTOPS_ALERT_EMAIL = "ops@example.com"
$env:AGENTOPS_SMTP_HOST = "smtp.example.com"
# $env:AGENTOPS_SMTP_PORT = "587"
# $env:AGENTOPS_SMTP_USER = "..."
# $env:AGENTOPS_SMTP_PASSWORD = "..."
# $env:AGENTOPS_SMTP_FROM = "agentops@example.com"
# $env:AGENTOPS_SMTP_TLS = "1"
agentops-alerts --db .\data\agentops.db

有告警时退出码 2。--dry-run 只打印,不 webhook / 不发信。

事件 outbox / Kafka

$env:AGENTOPS_EVENT_OUTBOX = ".\data\events.jsonl"
# 可选:同时发到 Kafka(需 pip install "agent-operations[kafka]")
$env:AGENTOPS_KAFKA_BOOTSTRAP = "localhost:9092"
# $env:AGENTOPS_KAFKA_TOPIC = "agentops.run"          # 默认
# $env:AGENTOPS_KAFKA_CLIENT_ID = "agentops"          # 默认

每次成功写入追加一行 agentops.run v1 事件(create / finalize);若配置了 bootstrap 则再 publish 同一 JSON。

另开进程消费写入本地库(不会再次 emit,避免回环):

# $env:AGENTOPS_KAFKA_GROUP = "agentops-ingest"   # 默认
agentops-kafka --db .\data\agentops.db
# agentops-kafka --max-messages 100

API Key(外网绑定前必配)

$env:AGENTOPS_API_KEYS = '{"dev-key":{"projects":["demo"]},"admin":{"projects":["*"]}}'
agentops-http --host 0.0.0.0 --port 8787

请求头:Authorization: Bearer <key> 或 X-API-Key: <key>。
未配置密钥时:仅建议本机 127.0.0.1,接口开放。非 loopback 绑定必须配置密钥。

调试页有 API key 输入框(存浏览器 localStorage),Reload 时自动带 X-API-Key。

限制(请先读)

限制 说明
Alpha 接口可能继续演进;生产使用请自行评估
DB 默认 SQLite;也可 connect / agentops-http --db / AGENTOPS_DATABASE_URL 用 postgresql://…(需 [postgres])
自报数据 延迟 / token / 费用依赖调用方或估算,不是云厂商账单
非全链路 APM 聚焦 agent run + steps,不是通用分布式 tracing
Kafka JSONL outbox + 可选 producer / agentops-kafka consumer;无内置 broker
页面 只读调试页;可填 API key 后查询
安装渠道 PyPI:pip install agent-operations;版本见 PyPI / tag v0.1.5

更细的后续项见 ROADMAP.md。

开发

推荐用 uv 虚拟环境(本机已可用 uv):

uv venv .venv
.\.venv\Scripts\Activate.ps1
uv pip install -e ".[agentic-loop,openai,langgraph]"
uv pip install build twine   # 仅发版需要
python -m unittest discover -s tests -v
agentops-http

验证 PyPI 包时另建干净环境,例如:

uv venv .venv-release --clear
.\.venv-release\Scripts\Activate.ps1
uv pip install "agent-operations==0.1.5"

库入口常用符号:connect, ingest_run, list_runs, summarize_runs, get_run, list_tool_failures, TracedRun。

Metadata

Release files for agent-operations 0.1.5

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

Source distribution (sdist)

Source distribution for agent-operations 0.1.5
File Size Uploaded Provenance
agent_operations-0.1.5.tar.gz 43.8 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for agent-operations 0.1.5
File Interpreter ABI Platform
agent_operations-0.1.5-py3-none-any.whl Python 3 none any Details

Total release size: 83.3 kB

Release files / agent_operations-0.1.5.tar.gz

Download URL agent_operations-0.1.5.tar.gz
Size 43.8 kB
Tags Source
SHA-256 checksum
How to use checksums
6ab15fb47200d2ac37804379eb94cadeff77cdd5906d820f6766abbb31212bde
BLAKE2b-256 checksum
How to use checksums
b8ce5b37e7e2b6dc623cf8fe3eb4beff9dbcfccb697b89829c057550b5e277fa
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.14.2

Release files / agent_operations-0.1.5-py3-none-any.whl

Download URL agent_operations-0.1.5-py3-none-any.whl
Size 39.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
2fa4d1d212cf36a58a24a5fec3e1992940002feea5a4eaad1357adf989d70f97
BLAKE2b-256 checksum
How to use checksums
452bbf1c5eef2409a9d44ee97dacd66fcc9045944ae2e11df15dc7b47abcd600
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.14.2

Security for agent-operations 0.1.5

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.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

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