Partnership outreach
What it does
For a podcast, channel, or brand the market research agent found, it drafts a short first message proposing a collaboration that clearly helps both sides, such as a guest-episode swap.
See Leads → Partners.
What it reads
- What is known about the partner from the market research: its channel, its audience, and its angle.
- Your diagnosis: your positioning, audience, and offer.
What it produces
A draft in four labelled parts:
- Subject: one line.
- Value for them: why the collaboration helps their show and audience.
- Value for us: why it helps you.
- Message: a 120–180 word body ending with one small ask, such as a reply to schedule a recording.
Human oversight
- It only drafts. Each message is saved with status
draft. You review it, edit it, and send it yourself. PULSE sends nothing.
Model
The deployment’s base model, shared with every other PULSE agent — on Runink Server, a Qwen-family model under the Apache-2.0 licence. See Sovereign inference.
Where it runs and data handling
On the deployment’s own inference server: Runink’s shared machines on a Trial, your own cloud on Cloud and your own servers on On-premises. No outside AI service is used on any plan.
Guardrails
- Two-sided value first. Every message says what the partner gets, not only what you want.
- No hard sell. No hype, no spammy superlatives, no pressure. The goal of a first message is a reply, not a deal.
- Never invents facts about the partner. When little is known about a show, it keeps its claims general rather than making up numbers. Requests to invent audience or download figures are refused.
- Draft only. Requests to send, publish, or pay are refused.
- The shared input guardrails apply to every request.
Getting good results
- It writes from what the market research found, so a fuller partner record gives a more specific message.
- It drafts the message; you send it and follow up on replies.