Skip to main content

AgentScope Python SDK

Project description

AgentScope Python SDK

Python SDK for AgentScope telemetry, tracing, and instrumentation.

Install

pip install agentscope-sdk

Quick Start

from agentscope import observe_run, observe_span, trace

with observe_run("coding_agent", agent_name="demo"):
    with observe_span("llm_call", provider="openai", model="gpt-4o"):
        trace.log("sending prompt", level="info")

Set AGENTSCOPE_API_BASE=http://localhost:8080 and AGENTSCOPE_API_KEY=... as needed.

Auto Instrumentation

from agentscope import auto_instrument

auto_instrument(["openai", "anthropic"])

This enables automatic provider tracing, orchestration boundary detection, and context propagation.

Default orchestration detectors:

  • HTTP ASGI (FastAPI/Starlette) request boundary
  • gRPC server interceptors
  • Celery task execution boundary
  • kafka-python consumer poll boundary (best-effort)

Default propagation:

  • HTTP headers via requests
  • Kafka message headers via kafka-python producer

Each boundary creates/continues one root orchestration run so agent runs inside that boundary share one trace_id automatically.

Disable orchestration auto-detection if needed:

auto_instrument(["openai"], orchestration=False)

Choose specific transports:

auto_instrument(["openai"], orchestration=["http", "grpc"])

Disable propagation injection:

auto_instrument(["openai"], propagation=False)

Coding-Agent Helpers

from agentscope import coding_agent_run, read_file, run_command, write_file

with coding_agent_run(agent_name="codebot"):
    content = read_file("README.md")
    write_file("notes.txt", content)
    result = run_command("echo done")

API

  • observe_run(...)
    • Cross-run linkage params: trace_id=..., parent_run_id=..., root_run_id=...
  • observe_span(...)
  • trace.auto(providers=None)
  • trace.log(message, level="info", span_id=None, metadata=None, timestamp=None)
  • trace.update_span(span_id, data)
  • auto_trace(providers=None)
  • auto_instrument(providers=None, orchestration="auto", propagation=True, orchestration_workflow_name="orchestration_auto_trace", orchestration_agent_name="orchestrator")
  • coding_agent_run(agent_name="coding_agent")
  • instrument_coding_agent(fn)
  • read_file(path, encoding="utf-8")
  • write_file(path, content, encoding="utf-8")
  • run_command(command, cwd=None, env=None, check=False, shell=None)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

agentscope_sdk-0.1.13.tar.gz (21.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

agentscope_sdk-0.1.13-py3-none-any.whl (30.2 kB view details)

Uploaded Python 3

File details

Details for the file agentscope_sdk-0.1.13.tar.gz.

File metadata

  • Download URL: agentscope_sdk-0.1.13.tar.gz
  • Upload date:
  • Size: 21.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.2

File hashes

Hashes for agentscope_sdk-0.1.13.tar.gz
Algorithm Hash digest
SHA256 8b2147a6cb95d218041c3f341ba673fe76d417092fbb60f18d92f59152236a77
MD5 cb993963957343d29b528a1213698064
BLAKE2b-256 394e563c55983b246b076382b9e2fc194b859acacefe34fe09581b0489fb9ee8

See more details on using hashes here.

File details

Details for the file agentscope_sdk-0.1.13-py3-none-any.whl.

File metadata

File hashes

Hashes for agentscope_sdk-0.1.13-py3-none-any.whl
Algorithm Hash digest
SHA256 112367c88984d5d0830e38750fc7f569a925a1442f8c49ed1d33aeee9bc1eb0a
MD5 58cb79f98d09753f69060790731de4e5
BLAKE2b-256 102f4ab78cbc766156988d80a9a2859cc034d953e5076a67fe39cba69d3d5e9a

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page