Agents In Use
What’s new todayLast update
108 use cases added today

Use Claude to file home reports in Obsidian

Wes Kennedy uses Claude Code to maintain an Obsidian vault whose notes are ordinary Markdown files. A root CLAUDE.md describes naming, tags, folder structure and daily routines, providing the context for handling material that arrives in an inbox folder.

Source published wes.todayAgent: Claude
OUR ADAPTATION

Suggested prompt

Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.

Read new home reports in my Obsidian inbox. Convert them into named Markdown notes with tags and metadata using my vault conventions, file them under the appropriate home project, and summarize what changed.

Evidence at a glance

Sourced account
Source author
Wes Kennedy
Original account
Read the source ↗
Published
Source link checked
Reported result
Roof-measurement and solar structural reports filed as named, tagged notes in Personal/Home/, alongside contextual home-project material.
Human involvement
Kennedy maintained and corrected the workflow; no separate approval step for each home report is described.

A linked account supports attribution. It does not mean we independently reproduced its result.

Report an issue with this caseOpen correction email

Opens your email app with the case link included. Nothing is sent automatically.

How the agent did it

  1. Define folder, naming and tagging conventions in the vault instructions.
  2. Drop home reports into the inbox for the check-out routine.
  3. Inspect the named, tagged notes and day's processing summary.

Wes Kennedy uses Claude Code to maintain an Obsidian vault whose notes are ordinary Markdown files. A root CLAUDE.md describes naming, tags, folder structure and daily routines, providing the context for handling material that arrives in an inbox folder.

During daily check-out, Claude scans that inbox. It reads newly added PDFs, CSVs, transcripts or downloaded files, converts the content into Markdown with metadata and tags, and files it in the appropriate location. It then mentions what it processed in the day's summary. Kennedy's concrete home example was a roof-measurement PDF from an insurance report alongside a structural-engineering report for solar installation. Both became named, tagged documents under Personal/Home/.

The same broader system also processed an outdoor-remodel voice memo into sections, unresolved questions and next steps. Kennedy describes that memo as a rambling discussion of retaining walls, irrigation, porches and city-code research, illustrating why the home reports belonged beside contextual notes rather than in a disconnected download pile.

He emphasizes that the setup took weeks of iteration. Instructions drifted, and the voice pipeline had repeated-processing and iCloud-access bugs. He edits drafts and confirms proposed tasks. The article describes filing the home reports but no separate document-by-document approval checkpoint.

Prompt
Suggested prompt (written by Agents In Use, not the author's)

Source: wes.today · Wes Kennedy

Frequently asked questions

What made Obsidian accessible to the agent?

The vault is a directory of Markdown files, so Claude Code can read and edit it through the filesystem.

Which home reports were actually processed?

Kennedy names a roof-measurement insurance PDF and a structural-engineering report for solar-panel installation.

Was the setup reliable without iteration?

No. Kennedy reports instruction drift, repeated voice processing and iCloud-access bugs during weeks of refinement.

More details

About this account

Claude read inbox files, converted them to tagged Markdown and selected their vault destination.

Firsthand source opened and read. Suggested prompt written by us from the reported workflow; no original task prompt is quoted.

Credit: Wes Kennedy

Source published

Added to Agents In Use

Source checked

Tools and topics

Claude Code · Obsidian · CLAUDE.md · Local PDFs

Home documents · Obsidian inbox · Report filing

Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.