logspine 0.1.0

Drop-in observability for AI agents — Python SDK

logspine · Python SDK

Drop-in observability for AI agents. Track every OpenAI and Anthropic call with cost, tokens, and latency — no code changes to your prompts.

Full docs: logspine.dev/docs

---

Install

pip install logspine

# with OpenAI support
pip install "logspine[openai]"

# with Anthropic support
pip install "logspine[anthropic]"

# both
pip install "logspine[openai,anthropic]"

---

Quick start — OpenAI

import openai
from logspine import LogspineClient

logspine = LogspineClient(api_key="lsk_live_...")
client = logspine.instrument_openai(openai.OpenAI())

# Every call is now tracked automatically
response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "Say hello in five words."}],
)
print(response.choices[0].message.content)

logspine.flush()  # call before process exit

---

Quick start — Anthropic

import anthropic
from logspine import LogspineClient

logspine = LogspineClient(api_key="lsk_live_...")
client = logspine.instrument_anthropic(anthropic.Anthropic())

response = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=256,
    messages=[{"role": "user", "content": "Say hello in five words."}],
)
print(response.content[0].text)

logspine.flush()

---

Optional span metadata

Pass a logspine dict as an extra kwarg to tag individual calls:

client.chat.completions.create(
    model="gpt-4o",
    messages=[...],
    logspine={
        "trace_id": "my-session-abc123",
        "name": "summarise_document",
        "user_id": "usr_42",
    },
)

---

Links

• Docs
• Dashboard
• Pricing
• GitHub