lookspan — local-first observability for AI agents

PyPI License: MIT Docs

Python SDK for Lookspan — emit spans from your AI agents to a dashboard that runs on your own machine. No cloud, no account, no API key, no data leaving your laptop.

If you have ever asked "why did my agent call that tool twice?" or "where did those 40 seconds go?", this is the missing view: every LLM call, tool call, chain step and MCP request as a span on a timeline you own.

  • Local-first. The dashboard is a process on 127.0.0.1. Your prompts stay yours.
  • Framework agnostic. Works with LangChain, LangGraph, CrewAI, MCP clients or plain code.
  • Content off by default. Prompts and completions are not recorded unless you opt in.
  • Small. One dependency (httpx), typed, MIT.

Install

pip install lookspan

The dashboard itself ships as a separate CLI:

npx lookspan-cli

Quick start

from lookspan import LookspanClient, Span, SpanType, SpanStatus

client = LookspanClient(endpoint="http://127.0.0.1:3100/api/ingest")

trace_id = client.new_trace_id()
span_id = client.new_span_id()

client.send([
    Span(
        trace_id=trace_id,
        span_id=span_id,
        parent_span_id=None,
        type=SpanType.LLM_CALL,
        name="my-agent.completion",
        started_at="2026-05-08T12:00:00Z",
        ended_at="2026-05-08T12:00:01Z",
        status=SpanStatus.OK,
        framework="custom",
        model="claude-opus-4-7",
        provider="anthropic",
    )
])

client.flush()

LookspanClient omits span input and output by default. Pass capture_content=True only when that content is safe to persist on disk.

Framework adapters

You rarely need to build spans by hand — install the adapter for your stack instead:

Package Traces
lookspan-langgraph LangChain chains, LangGraph nodes, LLM calls, tools
lookspan-crewai Crew kickoffs, agents, tasks, tools, LLM calls

Links

MIT licensed. Part of Fervon.

Project links

Data verified by PyPI on Aug 19, 2026
Data provided by the project maintainers, verified at the time the release was uploaded to PyPI.

Key dates

PyPI data
Data sourced directly from PyPI's database.
  • Released:
Latest release

1 maintainer

PyPI data
Data sourced directly from PyPI's database.
Avatar for jonathanMartin from gravatar.com jonathanMartin

Credits

Author/Maintainer: Jonathan Martin

GitHub Statistics

Data verified by PyPI on Aug 19, 2026
The GitHub source repository was provided by the project maintainers and verified by PyPI at the time of upload. Stars, forks, and open issues/PRs are derived from that repository and have not been independently verified.

License

MIT License (MIT)

Requires

Python >=3.10

Provides Extra

dev

Tags

agents ai anthropic crewai debugging instrumentation langchain langgraph llm llmops local-first lookspan mcp model-context-protocol monitoring observability openai self-hosted spans telemetry tracing

Classifiers

Report project as malware

Download files

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

Source Distribution

lookspan-0.5.3.tar.gz (6.9 kB view details)

Uploaded Source

Built Distribution

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

lookspan-0.5.3-py3-none-any.whl (6.9 kB view details)

Uploaded Python 3

File details

Details for the file lookspan-0.5.3.tar.gz.

File metadata

  • Download URL: lookspan-0.5.3.tar.gz
  • Upload date:
  • Size: 6.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for lookspan-0.5.3.tar.gz
Algorithm Hash digest
SHA256 0c932e7711e1a8fa437cb74ca369ec91d659ab65926818498f85674a25c8f339
MD5 6f705290eafd8cbe4359302308d8baea
BLAKE2b-256 60e4121268e12608eda000f9a3987004609eb3785a56dc75cd2cc5821899c9cb

See more details on using hashes here.

Provenance

The following attestation bundles were made for lookspan-0.5.3.tar.gz:

Publisher: publish-pypi.yml on JoniMartin27/lookspan

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lookspan-0.5.3-py3-none-any.whl.

File metadata

  • Download URL: lookspan-0.5.3-py3-none-any.whl
  • Upload date:
  • Size: 6.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for lookspan-0.5.3-py3-none-any.whl
Algorithm Hash digest
SHA256 a63ca0c0f7de8b36075a49a21dc48c4ed04b26da3f21defd7ccf8714ffd257e3
MD5 cb8b57513f57146eb0feb5a2e4321a5a
BLAKE2b-256 5751bf3c2a6a9e41553dfc9cd35c94d0f5fffa3844ab8485ab538eea9bf96914

See more details on using hashes here.

Provenance

The following attestation bundles were made for lookspan-0.5.3-py3-none-any.whl:

Publisher: publish-pypi.yml on JoniMartin27/lookspan

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

This release

0.5.3 This release

2 files

0.1.1

2 files

0.1.0

2 files

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