agentleFS
AndyMelnykovAndyMelnykov/obsidian-claude/AGENTS.md

obsidian-claude

AGENTS.md2 starsChanged 3 months agoMITView on GitHub

claude-obsidian is capture, organize, ask, defuddle, autoresearch, and bulk-import — six plain-Markdown Claude Code skills at skills/<name>/SKILL.md, with no Python core, no transaction bundles, no ledgers, no MoC maintenance. capture, organize, and ask need no network access; defuddle and autoresearch require explicit, per-request consent before any egress, and defuddle additionally requires an external Defuddle-style extractor the user provides. bulk-import files many sources in one run, reusing capture's and defuddle's rules with a single upfront consent for any URLs in the…

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,439 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository6 other files it gives its agents.

The file

We split this AGENTS.md into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No product/vault distinction, no marker file, no registration step.

    line 23
  2. Very specific:

    Read the skill completely from skills/<name>/SKILL.md.

    line 27
  3. Distinctive:

    Read skills/capture/references/note-format.md — all five skills share this note format, folder-placement heuristic, slug-uniqueness rule, …

    line 28

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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
AGENTS.md

claude-obsidian: Agent Instructions

claude-obsidian is capture, organize, ask, defuddle, autoresearch, and bulk-import — six plain-Markdown Claude Code skills at skills/<name>/SKILL.md, with no Python core, no transaction bundles, no ledgers, no MoC maintenance. capture, organize, and ask need no network access; defuddle and autoresearch require explicit, per-request consent before any egress, and defuddle additionally requires an external Defuddle-style extractor the user provides. bulk-import files many sources in one run, reusing capture's and defuddle's rules with a single upfront consent for any URLs in the batch rather than one per source. Notes are filed under notes/<topic>/ or projects/<project-slug>/ (never at the vault root directly), and indexes/<Topic>.md pages give a fast entry point into each notes/ topic. See docs/superpowers/specs/2026-08-22-simple-capture-design.md for the design and skills/capture/references/note-format.md for the note format all five skills share.

Primary system: capture / organize / ask / defuddle / autoresearch / bulk-import

No product/vault distinction, no marker file, no registration step. The user just points Claude Code at whatever directory they treat as their vault (opened in Obsidian or not) and invokes a skill.

  1. Read the skill completely from skills/<name>/SKILL.md.
  2. Read skills/capture/references/note-format.md — all five skills share this note format, folder-placement heuristic, slug-uniqueness rule, and linking pass. autoresearch additionally reads its own program.md for research budget and evidence-discipline defaults.
  3. Resolve the vault: an explicit directory named in the request, then the OBSIDIAN_VAULT environment variable if set, then the current directory. Set OBSIDIAN_VAULT once to use ask/capture/etc. from any other project's Claude Code or Codex session without repeating the path. capture creates inbox/, notes/, projects/, indexes/, templates/, and attachments/ in the resolved vault on first use if any don't exist.

Behavior, triggers, and out-of-scope boundaries for each skill are fully specified in its own SKILL.md — this file doesn't duplicate them. organize never runs automatically; ask never writes to the vault; defuddle and autoresearch never fetch or research without explicit per-request consent, and never file a note without a separate, later consent to keep the result. bulk-import only runs when handed more than one source, gets one upfront consent for every URL in its batch instead of one per source, and skips/defaults collisions and ambiguous placement rather than asking mid-batch. None of the six uses transactions, ledgers, or an index/MoC.

Reference

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.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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.