v0.1.0 — Initial release

First public release of tiktok-metadata-kit.

What's included

• Research API client (tiktok_metadata_kit.research_api.ResearchAPIClient)
  • Automatic OAuth client-credentials token retrieval and proactive refresh
  • Single-page primitives: query_videos, query_user_videos
  • Streaming generators with cursor-based pagination: iter_videos,
    iter_user_videos, iter_video_pages, iter_user_video_pages
  • Typed query parameters via QueryOptions / PageOptions TypedDicts
  • Configurable response fields (constructor default + per-call override)
  • Retry with exponential backoff on transient failures (429, 5xx, network
    errors); honors Retry-After
  • Sync httpx client with connection pooling and context-manager support
• Python 3.12+, fully typed (py.typed marker shipped)

Full Changelog: https://github.com/Nico-AP/tiktok-metadata-kit/commits/v0.1.0

Not yet included

The scraper subpackage is stubbed but not functional. The pip install "tiktok-metadata-kit[scraper]" extra installs its runtime dependencies in anticipation of a future release.

Installation

pip install tiktok-metadata-kit

Quick start

from tiktok_metadata_kit.research_api import ResearchAPIClient

with ResearchAPIClient("client_key", "client_secret") as client:
    for video in client.iter_user_videos(["username"]):
        print(video["id"], video["view_count"])

See the README for filter options, pagination control, and field selection.

Requirements

• Python ≥ 3.12
• TikTok Research API credentials (apply at the TikTok developer portal)