All flashcards — learn from mistakes

Updating context conversationally: telling the agent "save this," "remember," or "you should always" triggers it to edit the context file on your behalf. No special syntax needed.

Each new chat loads the current context file at startup. Changes propagate to the next session immediately — no restart, no caching, no opt-in for users.

A reusable recipe: name a markdown file, declare its purpose in the context file, tell the agent when to append. Generalizes to any side effect — logs, decisions, overrides.

The agent's behavior when it cannot resolve a question: attempt a partial answer, narrate the failure in chat, and append the question to the audit file — all in one response.

A plain markdown document in the site's agent assets, loaded at the start of every session. Editable like any other doc — no deploy step. Carries owners, processes, conventions, and decay rules.

The behavior pattern enabled by learning rules. Correct the agent in chat; it extracts the underlying rule, saves it to context, and re-runs the query with the new knowledge.

SharePoint AI's context belongs to the site or team, not the individual user. Every chat session anyone starts on that site loads the same rules — making context a governance surface, not a personal setting.

A meta-instruction in the context file telling the agent to treat user corrections as permanent lessons and write them back to context. One sentence converts the system from static to self-improving.

Corrections written back into the context file as plain-text entries. Survive across sessions, inspectable by anyone, editable like any other markdown — your team's agent gets smarter without code.

A context-file rule that makes the agent append every unresolvable question to an audit markdown file. Turns the agent into a self-auditor without code — just a sentence of markdown.