bench-observe 0.1.0

Bench — the home for your AI agents. One line of code → live dashboard, public profile, README badge.

bench-sdk (Python)

Python SDK for Bench — the home for your AI agents.

pip install bench-observe

Usage

import asyncio
from bench import observe

agent = observe(api_key="bk_xxx", agent="my-agent")

async def search(t):
    result = await do_search()
    t.log("found", len(result))
    t.cost(0.004)
    return result

async def main():
    result = await agent.task("search", {"query": "python"}, search)

asyncio.run(main())

Context manager

async with agent.task_ctx("summarize", {"doc": url}) as t:
    summary = await llm.generate(prompt)
    t.log("tokens", 1200)
    t.cost(0.012)

Subtasks

async def pipeline(t):
    await t.subtask("fetch", {"url": url}, fetch_fn)
    await t.subtask("parse", {"html": html}, parse_fn)
    await t.subtask("embed", {"text": text}, embed_fn)

await agent.task("full-pipeline", input, pipeline)

Custom events

agent.event("tool.use", {"tool": "search", "query": q})

LangChain adapter

pip install bench-langchain

from bench import observe
from bench_langchain import BenchCallbackHandler

agent = observe(api_key="bk_xxx", agent="my-langchain-agent")
handler = BenchCallbackHandler(agent)

# Every chain/LLM/tool call is now a traced Bench task
chain.invoke({"input": "..."}, config={"callbacks": [handler]})

After one run you get a live dashboard, public profile, and README badge — no other setup.