agentleFS
nathanadaynathanaday/claude-atlas/skills/wiki-ingest/SKILL.md

wiki-ingest

Skill1 starsChanged 38 days agoMITView on GitHub

Triage what waits in ingest/, and capture what the wiki should learn from: each item goes to the wiki as a source, or to a note in scratchpad/ when it holds work or an idea. Use for ingest, process ingest/, process the inbox, add this file to the wiki, read and file this, batch ingest. Saving part of this conversation is wiki-save; an idea to act on later is a note in scratchpad/ (the almagest skill).

45 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use wiki-status, wiki-capture and 9 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • nexuschainsSkill

    wiki-status

    Reports the current state of the wiki vault — page counts, category breakdown, hub pages, orphans, stale entries, and a delta showing what has changed since the last ingest. Use this skill when the operator asks "what's the status", "wiki dashboard", "how many pages", "what's stale", "what's been ingested", "show me the delta", or "wiki health overview". Also surfaces structural insights (top hubs, fragmented clusters, bridge pages) when the operator asks "wiki insights", "what's central", or "show me the hubs". Read-only: this skill reports and recommends; it does not modify any files.

    Also in 5 projects that use this

  • jehrhardtSkill

    wiki-capture

    Quickly capture raw thoughts, links, images, or notes to today's Obsidian daily note from anywhere in Pi. Use when the user wants to jot something down for later wiki processing.

    Also in 6 projects that use this

  • Ivy-Chen1999Skill

    wiki-research

    Autonomously research a topic via multi-round web search, synthesize findings, and file structured results into the Obsidian wiki. Use this skill when the user says "/wiki-research [topic]", "research X", "find everything about Y", "do a deep dive on Z", "autonomous research on X", or wants comprehensive, web-sourced knowledge on a topic filed directly into their wiki.

    Also in 5 projects that use this

  • nexuschainsSkill

    cross-linker

    Scans the wiki vault and adds missing [[wikilinks]] between pages that are related but not yet connected. Use this skill when the operator says "link my pages", "find missing links", "cross-reference the wiki", "connect my wiki", "add wikilinks", or after any large ingest to weave new pages into the existing knowledge graph. Also trigger when orphaned pages need connecting or when wiki-lint reports fragmented tag clusters. This is a write-heavy skill — it modifies page files to insert links, unlike wiki-lint which only reports.

    Also in 4 projects that use this

  • Ivy-Chen1999Skill

    wiki-setup

    Initialize a new Obsidian wiki vault with the correct structure, special files, and configuration. Use this skill when the user wants to set up a new wiki from scratch, initialize the vault structure, create the .env file, or says things like "set up my wiki", "initialize obsidian", "create a new vault", "get started with the wiki". Also use when the user needs to reconfigure their existing vault or fix a broken setup.

    Also in 5 projects that use this

  • bozhouDevSkill

    wiki-init

    Use when bootstrapping a new personal wiki for any knowledge domain — research, codebase documentation, reading notes, competitive analysis, or any long-term knowledge accumulation project.

    Also in 7 projects that use this

  • +1 more from nexuschains
  • +3 more from Ivy-Chen1999

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 2 are distinctive, and 6 are very specific to this project.

  • 0

    Common

    Most of the 34 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 6

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Call vault for the files in ingest/, the tag list, and the vault's tagging mode.

    line 19
  2. Very specific:

    The work document: When the request names one (the palette's message says "Your work document is [[…]] (<id>)"), use its id.

    line 23
  3. Very specific:

    Call source with action: capture for the wiki items: ingest (the names), or text and title, and tags.

    line 43
  4. Very specific:

    Never move or capture a file from journals/: the journals are the user's.

    line 50
  5. Distinctive:

    Ask first only when the sources total more than about 600 pages.

    line 51
  6. Very specific:

    When no item goes to the wiki, end the work document with change action: reject and the reason ("every item went to scratchpad/"), and say …

    line 53
  7. Very specific:

    When progress or propose refuses with "the user cancelled ….

    line 56
  8. Distinctive:

    No step checks the vault's git state, and you never report it: every write tool commits the user's hand edits as a snapshot before it …

    line 59

“Common” is measured: kept by at least 6 in 10 of the 34 edited versions of this file we could read. “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

wiki-ingest

ingest/ holds files and notes the user dropped in. A paper teaches the wiki; a note that asks for work or holds an idea goes to scratchpad/. This skill sorts the items, captures the sources with their tags, moves the work notes, and hands the sources to wiki-sync, which writes the topics. One work document records the ingest from the first step to the decision; the user watches it in Obsidian and decides once, at the end.

Tools: vault, change (start, progress), source (capture). References: changes.md.

Procedure

  1. Call vault for the files in ingest/, the tag list, and the vault's tagging mode. For text the user pasted, use text and a title in step 5. A note with no type in tool/source-core/documents/ (lint's untyped) waits to be ingested too: move it into ingest/ first, because capture takes only names in ingest/.
  2. The work document. When the request names one (the palette's message says "Your work document is [[…]] (<id>)"), use its id. Otherwise call change with action: start, kind: ingest, and files (the names in ingest/ that the request takes). Report each later step with change action: progress, one line each.
  3. Triage each item:
    • a document to learn from (a paper, notes, an article, a design) → wiki;
    • a note that asks for work or holds an idea ("fix the login timeout", "idea: …") → scratchpad;
    • both → capture it, then write the work part in a note in scratchpad/ that links the new source;
    • a note that asks to read something later that is not in ingest/ → scratchpad. Give each wiki item its tags: the tags that exist and fit its subject. Use a tag that exists before you make a new one.
  4. Record the triage and go on; do not wait for a yes. Report it as one progress line ("triaged 5 items: 3 to the wiki, 2 to scratchpad/"), and keep one line per item (its destination and tags) for the change's notes. Ask the user only in these cases, with one question for every such item:
    • an item that fits no class above;
    • in tagging: known mode, an item that needs a tag no document holds.
  5. Call source with action: capture for the wiki items: ingest (the names), or text and title, and tags. One call takes one tag list, so put items with the same tags in one call. Set new_tags: true only when the user agreed to a new tag. A file captured before comes back with duplicate set and leaves ingest/. Report "captured <titles>".
  6. Move every scratchpad item from ingest/ to scratchpad/ with mv, and keep its name. Report "moved <names> to scratchpad/". Never move or capture a file from journals/: the journals are the user's.
  7. Ask first only when the sources total more than about 600 pages. Otherwise go on.

When no item goes to the wiki, end the work document with change action: reject and the reason ("every item went to scratchpad/"), and say so in one line.

When progress or propose refuses with "the user cancelled …; stop the work", stop at once and say in one line that the user cancelled the ingest.

No step checks the vault's git state, and you never report it: every write tool commits the user's hand edits as a snapshot before it writes.

Gate

None in the chat, except the questions of step 4. The topics wait for the user's decision in the work document, through the gate of wiki-sync.

Hand off

wiki-sync, with the captured source ids, the work document's id, and the triage lines for the change's notes.

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.