Name a topic, or none for overall, and your assistant gets the people your trust network relies on most for it, highest first, each with an optional one-line reason. Use it when you have no candidates yet. The answer repeats the topic UpTrust matched, so a short list reads as a real gap rather than a misunderstanding.
Ask it like this
- “Who should I read on nutrition science?”
- “Who does my field defer to on AI policy?”
When another tool fits better
- Which of these should I trust first?
You already have a list to put in order. - Who should I meet?
You want someone to talk to, not someone to read. - Who can help me with this?
You have a specific question you want answered.
Before you use it
- It needs the Read your trust signal permission, which UpTrust asks you to approve when you connect.
- It only reads. It changes nothing on UpTrust.
- In Claude, connect
https://mcp.uptrusting.com/mcp. In ChatGPT, connecthttps://mcp.uptrusting.com/chatgpt. How to connect. - It can also answer from another trust graph than yours, such as one built from academic papers; see which communities it can answer from.
What your assistant is told
Your assistant decides when to use this tool from the title and description UpTrust’s server gives it. Here they are exactly as the server sends them.
- Tool name
top_trusted_for_principal- Title
- Whose judgment is most trusted on a topic? (no list needed)
- Description
Use this when the user asks whose judgment to trust on a topic and has NO candidate list yet: "who should I read on X?", "who does my field defer to?", "trusted starting points for X", "who knows about X?". It returns the people the user's own trust network (the "principal" in the name is the signed-in user), or a named community when you route the call to one, trusts most on the topic, highest first, each with a user ID, a display name and an optional one-line reason; up to 50. This discovers candidates. Use is_trusted_by_principal instead to check one entity, rank_for_principal to sort a list the user already has, who_should_principal_meet for people to contact or be introduced to (this tool points at whose judgment to READ), and find_help when the user has a specific question they want answered. Omit `topic` for most-trusted overall. Topic is free text mapped to the issuer's taxonomy; the response echoes the resolved tag, so a thin result reads as a real trust gap rather than a taxonomy miss.