agentleFS
Sign inStart free
QuixiAIQuixiAI/agi-memory/skills/installed/personal-notes/SKILL.md

personal-notes

Skill617 starsChanged 10 months agoMITView on GitHub

Search, read, create, and carefully edit Markdown notes in a user-mounted Obsidian vault or Bear export

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 6 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask for the exact mounted vault or export path when it is not already explicit in the conversation or workspace configuration.

    line 18
  2. Very specific:

    Use a narrow glob, list_directory, or grep under that root.

    line 20
  3. Distinctive:

    Search titles, tags, links, and content, then read only the most relevant notes.

    line 24
  4. Very specific:

    Do not ingest an entire vault into memory by default.

    line 27
  5. Very specific:

    Prefer a targeted edit_file operation for an existing note.

    line 33
  6. Distinctive:

    Never overwrite a note to append a small section, never silently rename or move notes, and never resolve link conflicts by deleting content.

    line 34
  7. Distinctive:

    After a write, state the path and what changed.

    line 35
  8. Distinctive:

    If the vault is not mounted, tell the user to mount or copy it into the Hexis workspace and provide that path.

    line 39
  9. Distinctive:

    If writes are disallowed, provide the ready-to-paste Markdown and exact intended path so the work remains useful.

    line 41

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Personal Notes

Work with notes the user intentionally exposes inside the Hexis workspace. This supports Markdown-based Obsidian vaults and exported Bear notes; it does not reach into an unmounted home directory or Bear's private live database.

Establish the Note Root

  1. Ask for the exact mounted vault or export path when it is not already explicit in the conversation or workspace configuration.
  2. Confirm that the path is inside the permitted workspace. Never guess a path, scan the user's home directory, or reuse an unrelated repository because it happens to contain Markdown.
  3. Use a narrow glob, list_directory, or grep under that root. Expand the search only when the initial results justify it.

Read and Search

  • Search titles, tags, links, and content, then read only the most relevant notes.
  • Cite note paths in the answer so the user can locate the source.
  • Preserve the distinction between text in a note, a Hexis memory, and the assistant's inference.
  • Do not ingest an entire vault into memory by default. Use fast_ingest only for files the user explicitly chooses and explain that it creates derived memories.

Create and Edit

  1. Resolve the exact target path and inspect any existing file before proposing a change.
  2. Preserve YAML frontmatter, tags, wiki links, embeds, heading style, and surrounding formatting.
  3. Prefer a targeted edit_file operation for an existing note. For a new note, show the proposed path and content before write_file unless the user's request already specified both.
  4. Never overwrite a note to append a small section, never silently rename or move notes, and never resolve link conflicts by deleting content.
  5. After a write, state the path and what changed. If the tool reports a failure, do not claim the note was saved.

No Dead Ends

  • If the vault is not mounted, tell the user to mount or copy it into the Hexis workspace and provide that path.
  • For Bear, request a Markdown export or mounted export directory; do not advise modifying Bear's internal database.
  • If writes are disallowed, provide the ready-to-paste Markdown and exact intended path so the work remains useful.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free