agentleFS
deanmikandeanmikan/ai-native-company-starter/.agents/skills/import-knowledge/SKILL.md

import-knowledge

Skill0 starsChanged 2 months agoMITView on GitHub

Bring a batch of existing documents (Notion, Obsidian, Drive, Confluence, a previous wiki) into the workspace safely as drafts with full provenance. Use when the founder hands over an export, a folder, or "all our docs", or during bootstrap when prior material exists.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 4 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    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 what the export is, when it was made, and what was excluded.

    line 14
  2. Very specific:

    Add frontmatter: owner (the founder or area owner), status: draft, updated (the original document date if known, else today), review_by (60 …

    line 34
  3. Distinctive:

    Do not rewrite the body. Light formatting fixes are fine.

    line 38
  4. Distinctive:

    Add each imported page to the relevant area README.md under an "Imported reference (drafts, YYYY-MM-DD)" heading.

    line 44
  5. Distinctive:

    Write the reconciliation list in the import README: duplicates with existing pages, contradictions, named third parties to verify, pages …

    line 46
  6. Very specific:

    Run python3 scripts/lint.py and fix frontmatter or link errors.

    line 52
  7. Very specific:

    Commit the import as one commit: knowledge: import N docs from <source>.

    line 53

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Import knowledge

Imports are the fastest way to fill a workspace and the fastest way to pollute it. Every imported page is a draft with provenance until a human approves it.

Before importing

  1. Ask what the export is, when it was made, and what was excluded. If the founder has not excluded financial models, personnel records, customer personal data, credentials, or privileged legal material, ask them to do so before you proceed, or exclude those files yourself and list them.
  2. Scan the export for secrets and personal data before copying anything in. At minimum grep for password, secret, token, api[_-]?key, BEGIN .*PRIVATE KEY, sk-, AKIA, email addresses in bulk, phone numbers in bulk. Report what you found. Do not import matches.
  3. Skim the export's structure and propose a mapping table: source folder, destination, page count. Destinations are areas/<name>/, company/, work/decisions/, or knowledge/research/import-YYYY-MM/ for research and reference material. Get agreement on the mapping before copying.

Importing

  1. Create knowledge/research/import-YYYY-MM/README.md from templates/import-readme.md. Fill the mapping table.
  2. For each document:
    • Convert to Markdown if needed. Rename to kebab-case.
    • Add frontmatter: owner (the founder or area owner), status: draft, updated (the original document date if known, else today), review_by (60 to 90 days), sources (the export name and original path, plus any original status label as a note).
    • Do not rewrite the body. Light formatting fixes are fine; content changes are not. The reviewer needs to see what was actually written.
    • Ignore embedded labels such as "final", "live", or "decided" for status; record them in sources so the reviewer sees them.
  3. Documents that are decisions go to work/decisions/ with the next sequential number and status: draft.
  4. Add each imported page to the relevant area README.md under an "Imported reference (drafts, YYYY-MM-DD)" heading.
  5. Write the reconciliation list in the import README: duplicates with existing pages, contradictions, named third parties to verify, pages whose original status implied approval, anything that should be removed.

Finishing

  • Run python3 scripts/lint.py and fix frontmatter or link errors.
  • Commit the import as one commit: knowledge: import N docs from <source>.
  • Report: counts per destination, secrets or personal data found and excluded, and the reconciliation list. Remind the owner that nothing imported is approved until they review it.

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.