Your assistant gets the trust graphs it can query besides yours, such as ones built from academic papers or open-source code, and the named communities within each, with the tools each one supports. It checks this before answering as one of those communities. It reads no one’s trust graph.
Ask it like this
- “What perspectives can you show me?”
- “How would the open-source community rank these libraries?”
When another tool fits better
- What can I ask here?
You want to know what this connection can do in general.
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.
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
list_perspectives- Title
- Which communities and trust graphs can answer this?
- Description
Use this when the user asks which communities or trust graphs you can answer from ("what perspectives can you show me?", "how would that community see this?"), and before you answer as one. It lists the federated corpora (separate trust graphs a call can be routed to) and the named communities (perspectives) within each. Use what_can_i_ask instead when the user asks what this connection can do in general. Call this before setting `corpus`+`perspective` on a routable tool (rank, is_trusted, top_trusted, bridges, field — resolve_entity too, on corpora that implement it): each corpus's `routable_tools` in the answer is the authoritative list. It also returns each corpus's short name, what it is made of, and its community slugs. A corpus with an empty perspective list has not published its catalog yet — routing still works if you already know a slug. No trust graph is read; safe to call anytime.