docs: normalize Mintlify links

docs/agent.md

@@ -13,11 +13,11 @@ CLAWDBOT uses a single agent workspace directory (`agent.workspace`) as the agen
 
 Recommended: use `clawdbot setup` to create `~/.clawdbot/clawdbot.json` if missing and initialize the workspace files.
 
-Full workspace layout + backup guide: [`docs/agent-workspace.md`](https://docs.clawd.bot/agent-workspace)
+Full workspace layout + backup guide: [`docs/agent-workspace.md`](/agent-workspace)
 
 If `agent.sandbox` is enabled, non-main sessions can override this with
 per-session workspaces under `agent.sandbox.workspaceRoot` (see
-[`docs/configuration.md`](https://docs.clawd.bot/configuration)).
+[`docs/configuration.md`](/configuration)).
 
 ## Bootstrap files (injected)
 
@@ -52,7 +52,7 @@ Clawdbot loads skills from three locations (workspace wins on name conflict):
 - Managed/local: `~/.clawdbot/skills`
 - Workspace: `<workspace>/skills`
 
-Skills can be gated by config/env (see `skills` in [`docs/configuration.md`](https://docs.clawd.bot/configuration)).
+Skills can be gated by config/env (see `skills` in [`docs/configuration.md`](/configuration)).
 
 ## p-mono integration
 
@@ -91,7 +91,7 @@ message is injected before the next assistant response.
 
 When queue mode is `followup` or `collect`, inbound messages are held until the
 current turn ends, then a new agent turn starts with the queued payloads. See
-[`docs/queue.md`](https://docs.clawd.bot/queue) for mode + debounce/cap behavior.
+[`docs/queue.md`](/queue) for mode + debounce/cap behavior.
 
 Block streaming sends completed assistant blocks as soon as they finish; disable
 via `agent.blockStreamingDefault: "off"` if you only want the final response.
@@ -109,4 +109,4 @@ At minimum, set:
 
 ---
 
-*Next: [Group Chats](https://docs.clawd.bot/group-messages)* 🦞
+*Next: [Group Chats](/group-messages)* 🦞