# Harvest Notes

> Harvest tagged ideas from Obsidian daily notes. Extracts

- Skill: `markbaindesign/harvest-notes` (Agent Skill)
- Install (CLI): `npx skillmds@latest add markbaindesign/harvest-notes`
- Raw SKILL.md: https://api.skillmd.com/api/skills/markbaindesign/harvest-notes/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: markbaindesign (https://skillmd.com/u/markbaindesign)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/markbaindesign/harvest-notes

---


Run the Obsidian collector against unprocessed daily notes:

```
python3 studio/collectors/obsidian_collector.py
```

If the user passes `--all`, add that flag to reprocess all notes (including already-archived
ones in `processed/`).
If the user passes `--dry-run`, add that flag to preview without writing or moving files.

Side effects of a real (non-dry-run) run:
- Routes items to spec stubs / Ideas/ / feature backlog as before
- Records every extracted `#tag` (not just the routed ones) to
  `studio/collectors/obsidian_tagged_items.json` as structured data — one record per
  `{tag, text, date, source}`, deduplicated
- Moves each processed daily note into `processed/` inside the vault (lowercase — do not
  rename to `Processed/`, a differently-cased folder already exists from before this
  feature and collides on case-insensitive Dropbox sync)

After running, report:
- How many spec stubs were created and their names
- How many ideas were filed and to which topic files
- How many tagged items were recorded to the structured data file
- Any duplicates skipped

If new spec stubs were created, list them and ask if the user wants to open any for editing.