ingest

name	ingest
description	Ingest a document or URL into Kagura Memory — extract text, chunk it, and store an overview plus per-section memories. Use when the user wants to add a PDF, Office/HTML/EPUB file, audio/video, or a web page to their memory graph.

kagura ingest — document ingestion

Drive kagura ingest to turn a file or URL into a structured set of memories. Thin wrapper around the installed kagura CLI.

Preflight

• kagura --version; if missing → uv tool install kagura-memory (or pip install kagura-memory), then stop.
• Heavy parsers are opt-in extras (e.g. pip install 'kagura-memory[ingest-pdf]' or [ingest-all]); a missing parser surfaces a clear "install ..." error from the CLI — relay that hint rather than guessing.
• Requires authentication and a target context. Run kagura auth status; if not authed, run the auth skill first.

Run

kagura ingest ./report.pdf                  # local file
kagura ingest https://example.com/page      # URL
kagura ingest ./report.pdf --dry-run        # preview cost/sections; no LLM or backend write
kagura ingest ./report.pdf --json           # machine-readable result for scripting

Consume the result

• Relay the overview + per-section counts and any per-section error records.
• On a missing-extra error, surface the exact pip install command the CLI prints. Use --dry-run first when the user wants a cost/size estimate.