All flashcards — living knowledge base

An admin-only skill that generates a dated markdown report on KB health: entries, expert coverage, gap status, decay percentages, and recommended next actions. Governance through generation.

Discrete agent-callable capabilities the KB exposes (e.g. log gaps, ingest knowledge, find experts). The agent picks which to invoke from a user question; users do not call them by name.

The entry point to the KB. The agent grounds answers on the document library, cites the source file, and can silently fire side-effect skills like gap logging.

A markdown rulebook the agent loads at the start of every session. Defines storage conventions, decay handling, source-citation rules, and gap-routing — most "smart" behavior lives here.

The library of canonical markdown knowledge files. Files are produced by AI skills (not pasted by humans) and carry a last-verified date the agent reads on every query.

A SharePoint list mapping people to the domains and topics they own. The agent consults it to route gap items to the right human — routing is metadata, not inference.

The skill that promotes raw input into canonical entries. Reasons about what already exists, merges new info into the right files, updates verification dates, and resolves matching gaps.

A SharePoint site assembled from a document library, submissions folder, gaps list, expert directory, and context file — the substrate that AI Skills operate on as one organism.

Automatic freshness handling. Each file carries a last-verified date; the context file defines stale thresholds. Stale entries trigger verification tasks routed to owners.

A SharePoint list of questions the agent could not answer. The log-gaps skill creates items automatically; ingestion closes them when matching content lands. The system's memory of its own unknowns.

Standard SharePoint folder permissions applied to skill folders. Only users with access to the skill's folder can invoke it — AI governance reuses the existing identity model.

A staging location where users drop raw inputs (transcripts, emails, decks). Nothing enters the KB until an approver triggers the ingest skill — capture is decoupled from curation.

Ingestion captures the rationale behind a decision, not just the outcome. Best fed raw transcripts and email threads — the agent extracts both the number and the reasoning that produced it.