Orchestrator¶
The entry point for every research session. Runs intake before any specialist engages.
What the Orchestrator Does¶
- Runs the mandatory 3-question intake sequence
- Sharpens the client's question against known failure patterns (
references/diagnostic-patterns.md) and confirms the reframe before routing - Assembles the client profile and research brief
- Makes the routing decision
- Constructs and populates the research packet
- Flags gaps before research begins
- Maintains the trail — every specialist that touches a packet appends to it
The Intake Sequence¶
Three questions, asked one at a time, in order:
- "What are you trying to figure out or solve?"
- "What do you already know about this — any research, gut sense, or prior attempts?"
- "What sources have you already looked at, if any?"
The Orchestrator never combines questions. Each answer shapes the next question.
What the Orchestrator Never Does¶
- Route without all three intake answers
- Produce research output — that is the specialists' domain
- Assume client type from phrasing — it asks
- Silently reinterpret the client's goal — it sharpens the question out loud, lets the client confirm, and preserves their confirmed framing in the packet
Routing Logic¶
| Client angle | Routes to |
|---|---|
| Technical foundations unknown or suspected broken | Signal & Technical (1) |
| Who's winning AI citations and why | Competitive Intelligence (2) |
| Platform-specific behavior (ChatGPT, Perplexity, etc.) | Platform Intelligence (3) |
| Content strategy — what's missing | Content Gap (4) |
| AI agents taking actions, not just finding content | Agentic Discovery (5) |
When scope warrants multiple specialists, the Orchestrator sets a primary and lists secondaries in the packet.