Fleet mem

src/langsmith/fleet/essentials.mdx

@@ -91,19 +91,26 @@ To edit instructions:
 You can also update instructions by prompting the agent directly in the chat. For example: "Update your instructions to always respond in bullet points."
 </Tip>
 
-## Memory and updates
+## Memory
 
-Agents remember important information from previous conversations and can update themselves to work better.
+Agents remember important information from previous conversations and can update themselves to work better. Fleet agents use two sources of memory:
 
-- **Memory**: Agents persist relevant details from past interactions by writing files to a **memories folder** (using `write_file` and `edit_file` tool calls). This lets them make better decisions in future conversations.
-- **Self-updates**: Agents can add new tools, remove ones they don't need, or adjust their instructions to improve how they work.
-- **What stays the same**: Agents can't change their name, description, or the channels that start them.
+- **Thread-scoped memory**: Context from the current conversation thread, including messages and actions in that thread.
+- **Long-term memory**: Persistent files in the agent workspace, such as `AGENTS.md`, `tools.json` (tool configuration), `subagents/*`, and `skills/*`. These are loaded at runtime and available from the start of each run. `AGENTS.md` is inserted into the system prompt automatically. Other long-term files are not added to the prompt automatically; the agent must read them on demand (for example, using the `read_file` tool).
+
+Agents persist relevant details from past interactions by writing files to a **memories folder** (using `write_file` and `edit_file` tool calls). This helps them make better decisions in future conversations.
 
 <Note>
-By default, agents require approval before saving to the memories folder. You can disable this in the agent's settings. For agents that run on automated schedules (such as [schedules](/langsmith/fleet/schedules#add-a-schedule)), disable the approval requirement so the agent can persist information without manual intervention. See [Update memory](/langsmith/fleet/manage-agent-settings#update-memory) for instructions.
+By default, agents require approval before saving to the memories folder. You can disable this in the agent's settings.
+
+For agents that run on automated [schedules](/langsmith/fleet/schedules#add-a-schedule), we recommend [disabling the approval requirement](/langsmith/fleet/manage-agent-settings#disable-required-approval-for-memory-updates) so the agent can persist information without manual intervention.
 </Note>
 
-For more information, see [How we built Agent Builder's memory system](https://www.langchain.com/conceptual-guides/how-we-built-agent-builders-memory).
+For more information, see [How we built the memory system for Fleet (formerly known as Agent Builder)](https://www.langchain.com/conceptual-guides/how-we-built-agent-builders-memory).
+
+## Self-updates
+
+Agents can update themselves: they can add new tools, remove ones they don't need, or adjust their instructions. However, agents can't change their name, description, or the channels that start them.
 
 ## Skills
 

src/langsmith/fleet/manage-agent-settings.mdx

@@ -61,6 +61,8 @@ By default, your agent requires approval before saving to the memories folder. W
 If your agent runs on a [schedule](/langsmith/fleet/schedules#add-a-schedule) or other automated schedule, disable the memory approval requirement. Otherwise, the agent will pause on every scheduled run that involves a memory update and wait indefinitely for manual approval.
 </Tip>
 
+### Disable required approval for memory updates
+
 To disable the memory approval requirement:
 
 1. In the [LangSmith UI](https://smith.langchain.com), navigate to your agent's inbox.

src/langsmith/fleet/workspace-admin.mdx

@@ -4,14 +4,14 @@ sidebarTitle: Workspace admin
 description: Configure workspace-level settings for Fleet.
 ---
 
-This page covers workspace-level settings for Fleet, including secrets configuration and usage monitoring.
+Configure workspace secrets and manage spend limits for Fleet agents and users.
 
 ## Workspace secrets
 
 Fleet uses [workspace secrets](/langsmith/set-up-hierarchy#configure-workspace-settings) to store API keys for models and tools. The following secret types are available:
 
 - **Required model key**: An OpenAI or Anthropic API key is required for Fleet to make LLM API calls. The agent graphs load this key from workspace secrets for inference.
-- **Agent builder specific secrets**: Secrets prefixed with `FLEET_` are prioritized over workspace secrets within Fleet. This way, you can better track the usage of Fleet vs other parts of LangSmith which use the same secrets. If you have both `OPENAI_API_KEY` and `FLEET_OPENAI_API_KEY`, the `FLEET_OPENAI_API_KEY` secret will be used.
+- **Fleet-specific secrets**: Secrets prefixed with `FLEET_` are prioritized over workspace secrets within Fleet. This way, you can better track the usage of Fleet vs other parts of LangSmith that use the same secrets. If you have both `OPENAI_API_KEY` and `FLEET_OPENAI_API_KEY`, the `FLEET_OPENAI_API_KEY` secret will be used.
 - **Optional tool keys**: Add keys for any tools you enable. These are read from workspace secrets at runtime.
   - `EXA_API_KEY`: Required for Exa search tools (general web and LinkedIn profile search).
   - `TAVILY_API_KEY`: Required for Tavily web search.
@@ -37,7 +37,7 @@ To add a secret:
 
 ## Usage and spend limits
 
-The **Usage** page gives workspace admins visibility into Fleet spend and the ability to set spend limits for agents and users. This page is will only be visible to users with the `fleet:read-admin-config` permission.
+The **Usage** page gives workspace admins visibility into Fleet spend and the ability to set spend limits for agents and users. This page will only be visible to users with the `fleet:read-admin-config` permission.
 
 ### View current spend