noahbfreedman-cloud/ShelfAI

The document-ops layer for agent context. Abstracts, chunking, and a learning loop for skill files.

34
/ 100
Emerging

Applies semantic chunking and tiered abstracts to agent skill files, enabling selective loading based on task relevance rather than reading entire monolithic documents. Operates as a preprocessing layer that pairs with retrieval systems—files flow through ShelfAI's structuring pass before reaching QMD/Hermes/SuperMemory, gaining 60%+ token savings through chunking. Features a self-improving learning loop where agents extract operational insights from sessions and rewrite their own abstracts and file structure, creating feedback between usage patterns and organization.

Available on PyPI.

Maintenance 13 / 25
Adoption 3 / 25
Maturity 18 / 25
Community 0 / 25

How are scores calculated?

Stars

4

Forks

Language

Python

License

Apache-2.0

Last pushed

Mar 27, 2026

Commits (30d)

0

Dependencies

5

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/noahbfreedman-cloud/ShelfAI"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.