agentleFS
zrosenfieldzrosenfield/sharepoint-ai-skills/Skills/ingest-knowledge/SKILL.md

ingest-knowledge

Skill96 starsChanged 3 months agoMITView on GitHub

Extracts tacit knowledge from any input — typed text, transcripts, emails, documents, or Slack threads — and writes it into the Knowledge Base library as structured markdown. Triggers when someone shares knowledge directly, pastes content, or uploads a file to the Submissions library. Resolves matching open Knowledge Gaps items automatically.

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

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When someone shares knowledge — typed, pasted, or by uploading a file to the Submissions library — extract tacit knowledge and add it to …

    line 6
  2. Distinctive:

    Check the Knowledge Base library for an existing file on this topic.

    line 22
  3. Distinctive:

    If a file exists, read it.

    line 23
  4. Distinctive:

    If no file exists, create a new markdown file named with a short slug (e.g., vendor-contracts.md).

    line 24
  5. Distinctive:

    Always include the Source line (who, when, what format), Added date, and Last Verified date (set to today).

    line 30
  6. Distinctive:

    After ingesting, check the Knowledge Gaps list for Open or Assigned items this knowledge answers.

    line 31
  7. Distinctive:

    If the new knowledge updates an existing KB entry, also update that entry's Last Verified date to today.

    line 32
  8. Distinctive:

    When processing transcripts or meeting notes, preserve the conversational context that reveals reasoning.

    line 34

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

When someone shares knowledge — typed, pasted, or by uploading a file to the Submissions library — extract tacit knowledge and add it to the Knowledge Base.

What to Extract

Listen for more than facts. Actively look for:

  • Decisions and their reasoning ("we went with X because...")
  • Tradeoffs acknowledged ("the downside is..." / "we accepted that risk because...")
  • Failed approaches ("we tried X and it didn't work because...")
  • Changes over time ("we used to do X, then switched to Y when...")
  • Edge cases and exceptions ("that works except when..." / "watch out for...")
  • Unwritten rules ("technically the policy says X but in practice we...")
  • Lessons learned ("if I had to do it again..." / "the thing nobody tells you is...")

How to Structure It

  1. Check the Knowledge Base library for an existing file on this topic.
  2. If a file exists, read it. Add new knowledge under the right heading. If the new information updates, contradicts, or adds context to existing content, note the change — don't silently overwrite. Add a new subsection or append under the existing one.
  3. If no file exists, create a new markdown file named with a short slug (e.g., vendor-contracts.md).
  4. For each piece of knowledge, use the sections that apply:
    • Current State — what's true now
    • Rationale — why it's this way
    • History — what came before or changed
    • Nuance — edge cases, exceptions, gotchas
  5. Always include the Source line (who, when, what format), Added date, and Last Verified date (set to today).
  6. After ingesting, check the Knowledge Gaps list for Open or Assigned items this knowledge answers. Update matching items: set Status to Resolved, Resolved Date to today, KB File to the filename, and Source Type to match the input format (Transcript, Email, Document, Slack, or Typed).
  7. If the new knowledge updates an existing KB entry, also update that entry's Last Verified date to today.

When processing transcripts or meeting notes, preserve the conversational context that reveals reasoning. "Dana said we switched to CrowdStrike after the Meridian incident" is more valuable than "Pen testing vendor: CrowdStrike."

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.