Knowledge Source Toggle
Show what the AI is allowed to see
AI with access to multiple connectors — Drive, Slack, Notion, Email, Web — should make those sources visible and switchable. The user picks which sources are in scope before the run, sees how many hits each contributed, and can flip any source off when the answer should not see it. No silent aggregation.
Framing
The problem
Multi-source AI silently aggregates from every connector — users cannot tell what the answer actually saw, or stop it from seeing something it should not.
The pattern
Show every connector as a visible toggle, attribute every claim to a source, and let users scope sources per query.
Why chat breaks here
Chat hides retrieval inside the system; the user gets one merged answer with no surface to scope or audit the inputs.
Risks
Too many sources at once create noise; weak attribution can imply sources were used when they were merely consulted.
Avoid when
The system has a single, well-known source and explicit toggling adds friction without insight.
Use when
Answers draw from multiple sources and users need to scope or audit which inputs were used.
DOPE evaluation
- Directability
- Toggle any source on or off per query, save scoped source-sets for repeat use
- Observability
- Every active source is visible, with hit counts after each run
- Predictability
- Removing a source removes its contribution — the answer changes accordingly
- Explainability
- Every claim in the answer is attributed to a specific source
In the wild
- Claude Projects Connectors (Anthropic) — Connectors launched July 2025 (50+ in directory by Feb 2026). Composer "+" → Connectors → toggle Gmail / Drive / Notion / Slack / GitHub / Asana per conversation. Toggles visible per session.
- NotebookLM Source Toggles (Google) — Per-source checkboxes in the Sources panel; queries can be scoped to a subset; citations attribute claims back to specific source spans.
- Perplexity Spaces Sources (Perplexity) — Per-Space toggle for Web vs My Files vs connectors, plus instructions field. The space is the container; the toggles are durable across threads inside it.
- ChatGPT Connectors (OpenAI) — Plus / Pro composer surfaces Gmail / Calendar / Drive / GitHub toggles per message. Per-message scope rather than per-conversation, but the toggle is visible and explicit.
FAQ
When should I use the Knowledge Source Toggle pattern?
Answers draw from multiple sources and users need to scope or audit which inputs were used.
When should I avoid the Knowledge Source Toggle pattern?
The system has a single, well-known source and explicit toggling adds friction without insight.
What problem does Knowledge Source Toggle solve?
Multi-source AI silently aggregates from every connector — users cannot tell what the answer actually saw, or stop it from seeing something it should not.
Why is chat the wrong fit for this?
Chat hides retrieval inside the system; the user gets one merged answer with no surface to scope or audit the inputs.
Related patterns
- Often paired with: Citation Trail — Pick which sources are in scope, then trace each claim back to the source that grounded it.
- Alternative to: Constraint Rules Engine — Soft scope per query vs hard rule across queries. Toggle is ephemeral, constraint is durable.
- Extends: Consent — Per-query source scope sits inside the durable consent grant — toggle narrows, consent permits.