Metadata-Version: 2.4
Name: traceshub
Version: 0.3.8
Summary: Capture AI coding sessions and sync them to Hugging Face Datasets — an open SpecStory.
Author: TracesHub contributors
License: MIT
Keywords: ai,claude,codex,coding-agents,datasets,huggingface,trajectories
Requires-Python: >=3.10
Requires-Dist: click>=8.1
Requires-Dist: huggingface-hub>=0.25
Requires-Dist: numpy>=1.26
Requires-Dist: pyarrow>=15.0
Requires-Dist: rich>=13.7
Provides-Extra: dev
Requires-Dist: pytest>=8.0; extra == 'dev'
Requires-Dist: ruff>=0.6; extra == 'dev'
Provides-Extra: embeddings
Requires-Dist: sentence-transformers>=3.0; extra == 'embeddings'
Description-Content-Type: text/markdown

# traceshub

Capture AI coding sessions (Claude Code, Codex CLI, Cursor, Gemini CLI, Aider)
into a local SQLite store and sync them to a Hugging Face Dataset.

```bash
uv tool install traceshub        # or: pipx install traceshub

traceshub init                       # create ~/.traceshub
traceshub login                      # sign in to Hugging Face
traceshub run claude                 # work normally; the session is captured
traceshub sync                       # push sessions to <user>/traceshub on HF
traceshub search "cuda memory leak"  # semantic search over your history
traceshub replay <session-id>        # replay a session in the terminal
```

Install real embeddings with `uv tool install "traceshub[embeddings]"`.
See [`../docs/SCHEMA.md`](../docs/SCHEMA.md) for the dataset layout.
