fix: quick updates

docs/v2/documentation/core-concepts/architecture.mdx

@@ -66,7 +66,7 @@ Every message is attributed to a specific peer and ordered chronologically withi
 
 ## System Components
 
-TODO: devs tell me if this section is legit or not pls 
+TODO: devs tell me if this section is legit or not pls
 
 
 At a high level, Honcho has three main components that work together.
@@ -91,15 +91,12 @@ The diagram above shows how agents write messages to Honcho, which triggers reas
 
 ## Perspective Taking
 
-One of Honcho's unique capabilities is modeling how different peers perceive each other based on their interactions. When multiple peers interact in a session, Honcho can build self-representations (a peer's representation built from all messages they've sent across all sessions) and other-representations (a peer's representation of another peer, built only from messages they've observed from that peer).
+One of Honcho's unique capabilities is modeling how different peers perceive each other based on their interactions. When multiple peers interact in a session, Honcho can build a representation of a peer itself (based on all the messages they've sent across all sessions) and representations of the peers a certain peer has interacted with (built only from messages they've observed from specific peers). It makes architecting the following figure effortless:
 
-TODO: Diagram here
+![](/images/perspectives.jpeg)
 
 This perspective-taking ability enables sophisticated multi-agent scenarios where each peer maintains distinct representations of the other participants they've interacted with.
 
-<Note>
-Perspective taking can be configured at the peer or session level depending on your application's needs.
-</Note>
 
 ## Configuration & Extensibility
 

docs/v2/documentation/core-concepts/reasoning.mdx

@@ -52,7 +52,7 @@ Here's an example of the data structure the reasoning models generate:
 }
 ```
 
-The reasoning models output their "thinking" followed by things that were explicitly stated, which serve as premises to scaffold deductive conclusions. It's this reasoning foundation that further reasoning is scaffolded. Currently that includes peer cards (personality summaries and psychological profiles), duplicate checking (identifying redundant or contradictory information), induction (pattern recognition across multiple messages), and abduction (inferring the simplest explanations for observed behavior).
+The reasoning models output their "thinking" followed by things that were explicitly stated, which serve as premises to scaffold deductive conclusions. It's on top of this reasoning foundation that further reasoning is scaffolded. Currently that includes peer cards (personality summaries and psychological profiles), duplicate checking (identifying redundant or contradictory information), induction (pattern recognition across multiple messages), and abduction (inferring the simplest explanations for observed behavior).
 
 The reasoning that Honcho does is something we're constantly iterating and improving on. Our goal is simple--provide the richest, most relevant context in the fastest, cheapest way possible in order to simulate statefulness in whatever setting you need.
 

docs/v2/guides/storing-data.mdx

@@ -58,4 +58,4 @@ In this case you can simply
 - Make a `Peer` for the AI
 
 Then you can make a `Session` for each thread of conversation and save
-`Messages` from the user and assistant in each turn of conversation
+`Messages` from the user and assistant in each turn of conversation