docs(open-tag): frame and link Claude Tag across README and docs

README.md

@@ -740,9 +740,10 @@ at MFS and focus on the app on top.
   or **plugins** on top of the same surface.
 
 One example is [`Open Tag`](examples/open-tag-skill/): a small open-source Slack
-bot you `@mention` — `@OpenClaude` or `@OpenCodex`, in the style of the official
-`@Claude` tag — that gathers authorized context and runs the task on a CLI agent
-backend. The point is how little it takes to build: MFS already gives it one
+bot you `@mention` — `@OpenClaude` or `@OpenCodex` — that gathers authorized
+context and runs the task on a CLI agent backend. It's an open, self-hosted homage
+to [Claude Tag](https://www.anthropic.com/news/introducing-claude-tag), Anthropic's
+hosted `@Claude` Slack teammate. The point is how little it takes to build: MFS already gives it one
 searchable, self-hosted memory over Slack, repos, docs, issues, and databases, so
 its strongest capability — plugging in whatever data sources you need — is also
 the easy part. See the [example](examples/open-tag-skill/) for the full walkthrough.

docs/integrations.md

@@ -8,7 +8,8 @@ There are two kinds of pages here:
 
 - **Apps on MFS** — runnable reference apps that use MFS as their searchable
   Memory.
-    - [Open Tag](integrations/open-tag.md) — a Claude Tag-style Slack tag-in bot.
+    - [Open Tag](integrations/open-tag.md) — a Slack tag-in bot, an open homage to
+      [Claude Tag](https://www.anthropic.com/news/introducing-claude-tag).
 - **Frameworks** — drop MFS into an existing agent / RAG stack.
     - [LangChain](integrations/langchain.md) — MFS as a `Retriever` and a tool.
     - [LangGraph](integrations/langgraph.md) — MFS as a retrieval node in a graph.

docs/integrations/open-tag.md

@@ -1,11 +1,18 @@
 # Open Tag — a Slack tag-in bot
 
-[Open Tag](https://github.com/zilliztech/mfs/tree/main/examples/open-tag-skill)
-is a small open-source demo of a **Claude Tag-style** Slack workflow built on
-MFS: you `@mention` a bot in a thread, it gathers authorized context, and a CLI
-agent backend does the work. By convention it answers to **`@OpenClaude`**
-(Claude backend) or **`@OpenCodex`** (Codex backend), so it reads like the
-official `@Claude` tag.
+[Claude Tag](https://www.anthropic.com/news/introducing-claude-tag) is Anthropic's
+way to bring Claude into Slack as a shared teammate: an admin grants it access to
+selected channels, tools, and data, and anyone in the channel can `@Claude` to
+delegate a task while they get on with other work. It is a hosted product for
+Claude Enterprise and Team.
+
+[Open Tag](https://github.com/zilliztech/mfs/tree/main/examples/open-tag-skill) is
+a small open-source homage to that pattern, built on MFS: you `@mention` a bot in
+a thread, it gathers authorized context, and a CLI agent backend does the work. By
+convention it answers to **`@OpenClaude`** (Claude backend) or **`@OpenCodex`**
+(Codex backend), so it reads like the official `@Claude` tag. The difference is
+that the **Memory** and **Tools** are your own self-hosted MFS sources, behind
+whatever CLI agent you run — your data and credentials stay on your machines.
 
 ## Architecture
 

examples/open-tag-skill/README.md

@@ -1,10 +1,17 @@
 # Open Tag Example
 
-Open Tag is a small open-source demo of a Claude Tag-style Slack workflow built on
-top of MFS: you mention a bot in Slack, the bot gathers authorized context, and a
-CLI agent backend does the work. The official tag bot answers to `@Claude` — by
-convention Open Tag answers to **`@OpenClaude`** (Claude backend) or
-**`@OpenCodex`** (Codex backend), so it reads the same way.
+[Claude Tag](https://www.anthropic.com/news/introducing-claude-tag) is Anthropic's
+way to bring Claude into Slack as a shared teammate: an admin grants it access to
+selected channels, tools, and data, and anyone in the channel can `@Claude` to
+delegate a task while they get on with other work. It is a hosted product for
+Claude Enterprise and Team.
+
+Open Tag is a small open-source homage to that pattern, built on top of MFS: you
+mention a bot in Slack, it gathers authorized context, and a CLI agent backend
+does the work. By convention it answers to **`@OpenClaude`** (Claude backend) or
+**`@OpenCodex`** (Codex backend), so it reads like the official `@Claude` tag. The
+difference is that the Memory and Tools are your own self-hosted MFS sources,
+behind whatever CLI agent you run — your data and credentials stay on your machines.
 
 The example maps the core concepts onto MFS: