Skip to main content

agentstream

Read, transform, and write AI agent session logs across formats (Claude, Codex, Gemini) with AG-UI events as the canonical format.

Installation

pip install agentstream

Quick Start

import agentstream

# Discover all agent sessions
sessions = agentstream.discover_sessions("~/myproject")

# Read as grouped messages
for msg in agentstream.read_messages(sessions[0]):
    print(f"{msg.role}: {msg.text[:100]}")
    for tc in msg.tool_calls:
        print(f"  -> {tc.name}({tc.args})")

Development

uv venv
uv pip install -e ".[dev]"

# Run tests with coverage
uv run pytest

# Type checking (ty is Astral's fast type checker)
uv run ty check

# Linting
uv run ruff check agentstream

Version Management

Python and Node.js versions are centralized in .python-version and .node-version files at the repo root.

Auto-updated from version files:

  • CI workflows
  • devcontainer
  • Docker builds (via make up in backend/)

Manual updates required when changing versions:

File Settings to update
pyproject.toml requires-python, target-version (ruff), python-version (ty)
daemon/pyproject.toml requires-python, target-version
backend/api/pyproject.toml requires-python, target-version

Version policy:

  • SDK (agentstream): Supports Python 3.12+ for broad compatibility
  • Backend services (daemon, api): Require Python 3.13+ (we control deployment)

License

MIT

Metadata

Release files for wandb-agentstream 1.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Built distribution (wheel)

Table of built distributions (wheels) for wandb-agentstream 1.1.0
File Interpreter ABI Platform
wandb_agentstream-1.1.0-py3-none-any.whl Python 3 none any Details

Release files / wandb_agentstream-1.1.0-py3-none-any.whl

Download URL wandb_agentstream-1.1.0-py3-none-any.whl
Size 1.3 MB
Tags Python 3
SHA-256 checksum
How to use checksums
640aceae6d7d66290b79a2025de6a776f01031efa33202bc8080750628789d65
BLAKE2b-256 checksum
How to use checksums
0639bdbb0b6878a8d39a97175303b4b4eb171e329e075bc7ab3ee3bce48dffba
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 22, 2026.

Transparency log

Release history Release notifications | RSS feed

This release

1.1.0 This release

1 release file

1.0.8

1 release file

1.0.7

1 release file

1.0.6

1 release file

1.0.5

1 release file

1.0.4

1 release file

1.0.3

1 release file

1.0.2

1 release file

1.0.0

1 release file

0.7.14

1 release file

0.7.13

1 release file

0.7.12

1 release file

0.7.11

1 release file

0.7.8

1 release file

0.7.7

1 release file

0.7.6

1 release file

0.7.5

1 release file

0.7.4

1 release file

0.7.3

1 release file

0.7.2

1 release file

0.7.1

1 release file

0.7.0

1 release file

0.6.11

1 release file

0.6.9

1 release file

0.6.8

1 release file

0.6.7

1 release file

0.6.6

1 release file

0.6.5

1 release file

0.6.4

1 release file

0.6.3

1 release file

0.6.1

1 release file

0.6.0

1 release file

0.5.4

1 release file

0.5.3

1 release file

0.5.2

1 release file

0.5.1

1 release file

0.4.3

1 release file

0.4.0

1 release file

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