agentleFS
jsuchetjsuchet/jonathan-skills-lab/skills/brain-add/SKILL.md

brain-add

Skill1 starsChanged 5 months agoMITView on GitHub

Quick capture into the second brain. Accepts text, a file path, or a URL. Processes content, adds frontmatter and wiki-links, and files it in the appropriate category.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 7 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the file (supports .md, .txt, .pdf, images)

    line 34
  2. Very specific:

    Use display text where natural: [[freelancing|freelancer]]

    line 61
  3. Very specific:

    If strong matches found, add them to the related: frontmatter field

    line 65
  4. Very specific:

    Add a [[New Note Title]] link in any strongly related existing notes (append to their related field too)

    line 66
  5. Very specific:

    Add new entry in the appropriate category section: - [[note-name]](category/subcategory/note-name.md) — One-line summary.

    line 70
  6. Distinctive:

    Update the note count in the header and the section header count

    line 71
  7. Distinctive:

    Update "Last updated: YYYY-MM-DD"

    line 72
  8. Distinctive:

    Use Bash to append directly without reading:

    line 75
  9. Distinctive:

    Always read brain-config.md first — it's the source of truth for filing decisions.

    line 96
  10. Distinctive:

    If category is ambiguous, ask the user rather than guessing wrong.

    line 97
  11. Distinctive:

    Do not create duplicate notes.

    line 99

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

/brain-add <text | file | URL>

Quick capture tool. Processes input and files it as a structured note in the brain.

Steps

  1. Read brain-config.md (taxonomy, tags, filing signals) and brain-config-rules.md (processing rules, MOC rules).
  2. Detect input type from the argument:
    • Starts with http:// or https:// → URL
    • Starts with / or ~ or contains a file extension → File path
    • Everything else → Raw text
  3. Process by type:

    URL

    • Fetch the page content using WebFetch
    • Extract: title, author, publication date, main content
    • Summarize key points (3-5 bullets)
    • Extract notable quotes
    • Identify action items or takeaways
    • Determine category by matching content against filing signals in brain-config.md

    File path

    • Read the file (supports .md, .txt, .pdf, images)
    • For PDFs: extract and summarize each major section
    • For images: describe content, extract visible text
    • For text files: analyze structure and key content
    • Determine category from content + filing signals
    • Move original file to inbox/.processed/YYYY-MM-DD-original-filename

    Raw text

    • Analyze the text content
    • Determine if it's a thought, fact, how-to, reference, or idea
    • Structure it into appropriate note type
    • Determine category from content + filing signals
  4. Determine note type based on content:
    • Web content → article
    • Own thoughts/opinions → note
    • Facts/specs/data → reference
    • Creative concepts → idea
    • Step-by-step instructions → process or guide
  5. Select tags from the controlled vocabulary in brain-config.md. Use 2-5 tags per note. If the content introduces a genuinely new topic not covered by existing tags, add the new tag to brain-config.md and note the addition.
  6. Create the note file:
    • Path: brain/{category}/{subcategory}/{kebab-case-title}.md
    • Full YAML frontmatter (title, type, tags, source, created, updated, status: seed, category, related, summary)
    • Structured body with content organized by type template
    • Insert 3-8 inline [[wiki-links]] for key concepts mentioned in the text
    • Use display text where natural: [[freelancing|freelancer]]
  7. Check for connections:
    • Grep across brain/ for notes that share tags or reference the same [[topics]]
    • If strong matches found, add them to the related: frontmatter field
    • Add a [[New Note Title]] link in any strongly related existing notes (append to their related field too)
  8. Update index.md:
    • Read brain/index.md
    • Add new entry in the appropriate category section: - [[note-name]](category/subcategory/note-name.md) — One-line summary. Tags: tag1, tag2
    • Update the note count in the header and the section header count
    • Update "Last updated: YYYY-MM-DD"
  9. Append to log.md (do NOT read the file — append only):
    • Use Bash to append directly without reading:
      echo '## [YYYY-MM-DD] ingest | {note-title}
      Filed to: {category}/{subcategory}. Source: {source type}. {X} wiki-links created.
      ' >> brain/meta/log.md
  10. Update hot cache:
    • Rewrite brain/meta/hot.md with updated recent activity, brain shape, and current themes
    • Keep under ~500 words
  11. Confirm with:
    📥 Added to brain/{category}/{subcategory}/{filename}.md
    Tags: #tag1 #tag2 #tag3
    Links: [[topic1]], [[topic2]]
    Connected to: {count} existing notes

Important

  • Always read brain-config.md first — it's the source of truth for filing decisions.
  • If category is ambiguous, ask the user rather than guessing wrong.
  • For URLs, if WebFetch fails or returns minimal content, note this and create a seed note with the URL for manual review.
  • Do not create duplicate notes. Search for existing notes on the same topic first (by title, source URL, or key terms). If found, suggest updating the existing note instead.

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.