evolv-agents

Agent runs as e-volv Observer reads them. Every model call, tool call and agent run becomes a span with the OpenTelemetry GenAI attributes, sent through the e-volv-logs SDK you already have, so Observer's Agents pages show each run with its tokens, cost, session and prompt version.

pip install evolv-agents

import evolv_agents as agents
from openai import OpenAI

agents.init(key=os.environ["EVOLVE_LOGS_KEY"],
            url="https://api.e-volv.io/api/public/v1/logs", service="support-bot")
client = agents.instrument_openai(OpenAI())

with agents.session(thread_id, user=user_id), agents.agent("triage"):
    res = client.chat.completions.create(model="gpt-5-mini", messages=messages)
    with agents.tool("read_ticket"):
        ticket = tickets.read(ticket_id)

What it gives you

• init(content=False, **e_volv_logs_options): content=True records prompts, completions, tool arguments and results (off by default). e-volv Launch flags are off too, since a Berserker project key has no flags:read; pass flags= to turn them on for a key that does.
• agent(name, agent_id=, provider=, model=): the run, as a context manager.
• chat(model, provider=): one model call; call c.usage(input=, output=, cache_read=, cache_write=, response_model=) inside.
• tool(name): one tool call.
• session(session_id, user=) and prompt(name, version): tag every span inside.
• instrument_openai(client), instrument_anthropic(client): wrap the client in place; chat completions, the Responses API and messages, streams included.
• trace_mcp(session, server_name): every call_tool is a tool span.

Already on OpenTelemetry? You do not need this package: point your exporter at https://api.e-volv.io/api/public/v1/logs/otlp/v1/traces with the ingest key as Authorization and Observer reads the GenAI spans directly.