@letta-ai/letta-code 0.27.22 → 0.27.24

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@letta-ai/letta-code",
-  "version": "0.27.22",
+  "version": "0.27.24",
   "description": "Letta Code is a CLI tool for interacting with stateful Letta agents from the terminal.",
   "type": "module",
   "packageManager": "bun@1.3.0",

package/skills/context-doctor/SKILL.md

@@ -116,7 +116,8 @@ Review changes, then commit with a descriptive message:
 cd $MEMORY_DIR
 git status                # Review what changed before staging
 git add <specific files>  # Stage targeted paths — avoid blind `git add -A`
-git commit --author="<AGENT_NAME> <<ACTUAL_AGENT_ID>@letta.com>" -m "fix(doctor): <summary> 🏥
+author_name="${AGENT_NAME:-$AGENT_ID}"
+git commit --author="$author_name <$AGENT_ID@letta.com>" -m "fix(doctor): <summary> 🏥
 
 <identified issues and implemented solutions>"
 

package/skills/creating-mods/SKILL.md

@@ -155,7 +155,7 @@ Before finishing, verify:
 | `references/providers.md` | Adding a custom model/API provider for local agents |
 | `references/events.md` | Reacting to lifecycle/tool/turn events or transforming turns/tools |
 | `references/permissions.md` | Enforcing dynamic tool allow/ask/deny policy before approval/execution |
-| `references/ui.md` | Panels (including order-0 statusline) or `ui.panels` capability guards are involved |
+| `references/ui.md` | Panels (including order-0 statusline and order-1 dreaming indicator) or `ui.panels` capability guards are involved |
 | `references/plan-mode.md` | Recreating plan mode with commands, tools, events, permissions, and local state |
 | `references/analysis-mode.md` | Phrase-triggered diagnostic mode with turn reminders (simpler than plan-mode) |
 | `references/architecture.md` | Multiple capabilities, local state, cleanup, background model work, or non-trivial composition |

package/skills/creating-mods/references/ui.md

@@ -37,7 +37,8 @@ if (letta.capabilities.ui.panels) {
 
 `order` is a signed coordinate around the input:
 
-- `order > 0` — above the input, higher nearer the top (default `100`).
+- `order > 1` — additive panels above the input, higher nearer the top (default `100`).
+- `order === 1` — replaces the default dreaming/reflection indicator above the input. Use this only when intentionally overriding that row; otherwise use `order > 1`.
 - `order === 0` — the primary line just below the input, overriding the built-in `agent · model`. This is the statusline slot; use `customizing-statusline` for that work.
 - `order < 0` — stacks below the primary line, `-1` closest.
 
@@ -50,16 +51,57 @@ render(ctx: {
   width: number;
   agent: { id, name };
   model: { id, displayName, provider, reasoningEffort };
+  backgroundAgents: Array<{
+    type: string;
+    status: string;
+    durationMs: number;
+    agentId: string | null;
+  }>;
+  subagents: { list(): SubagentLifecycleItem[] };
   row(left, right, width): string;
   columns(parts: string[], width): string;
+  link(label: string, url: string): string;
   chalk: ChalkInstance;
 }): string | string[]
 ```
 
-`row`/`columns` are ANSI-aware, so chalk-colored segments align correctly.
+`row`/`columns` are ANSI-aware, so chalk-colored segments and OSC-8 links align correctly. Use `link(label, url)` when a compact label should hyperlink to a URL without rendering the whole URL.
 
 Close panels when they are transient, and close/replace long-lived panels from the activation disposer if reload should remove them.
 
+### Panel use case: dreaming indicator overrides
+
+When a user asks to change the "dreaming" UI/indicator, the reflection status above the input, or to add the full background-agent URL, use an `order: 1` panel replacement. The user should not need to know any internal row/component name.
+
+Checklist:
+
+- Open a panel with `order: 1`, not an additive panel.
+- Read active hidden background agents from `ctx.backgroundAgents`; filter by `status` (`pending`/`running`).
+- Use `agent.agentId` to build `https://app.letta.com/chat/${agent.agentId}`.
+- If the user asks for the full URL, render visible text; do not use `ctx.link()` because it hides the URL behind OSC-8.
+- If preserving animation, own the timer and call `panel.update()`; clean up timer and panel in the disposer.
+- Keep `render()` pure: do not call diagnostics or mutate external state from render.
+
+Critical shape:
+
+```ts
+const panel = letta.ui.openPanel({
+  id: "dreaming-url",
+  order: 1,
+  render(ctx) {
+    const agent = ctx.backgroundAgents.find(
+      (a) => a.status === "pending" || a.status === "running",
+    );
+    if (!agent) return "";
+
+    const url = agent.agentId
+      ? `https://app.letta.com/chat/${agent.agentId}`
+      : null;
+    // Render spinner/label/elapsed, plus visible URL if requested.
+  },
+});
+```
+
 ### Commands that open panels
 
 If a command's `run()` opens a panel, guard the command **registration** on `letta.capabilities.ui.panels` — not just the `openPanel` call:

package/skills/customizing-statusline/references/api.md

@@ -52,7 +52,8 @@ letta.ui.closePanel(id: string): void
 
 `order` is a signed coordinate around the input:
 
-- `order > 0` — above the input. Higher numbers render nearer the top. Default when omitted is `100`.
+- `order > 1` — additive panels above the input. Higher numbers render nearer the top. Default when omitted is `100`.
+- `order === 1` — replaces the default product-status row (currently dreaming/reflection status). Use this only when intentionally overriding that row; otherwise use `order > 1`.
 - `order === 0` — the primary line just below the input. Overrides the built-in `agent · model`. Use this for the statusline.
 - `order < 0` — stacks below the primary line. `-1` sits closest to it, more-negative lower.
 
@@ -65,13 +66,15 @@ render(ctx: {
   width: number;            // visible columns available to the panel
   agent: { id, name };      // live at render time
   model: { id, displayName, provider, reasoningEffort };
+  subagents: { list(): SubagentLifecycleItem[] };
   row(left, right, width): string;     // left + right, right-aligned, ANSI-aware
   columns(parts: string[], width): string; // spread parts evenly, ANSI-aware
+  link(label: string, url: string): string; // compact OSC-8 hyperlink helper
   chalk: ChalkInstance;     // color helper
 }): string | string[]
 ```
 
-- `row`/`columns` measure visible width with ANSI stripped, so chalk-colored segments align correctly.
+- `row`/`columns` measure visible width with ANSI stripped, so chalk-colored segments and links align correctly.
 - The host clips each line to `width` and caps total height; the mod owns layout within that.
 
 ## Render rules

package/skills/initializing-memory/SKILL.md

@@ -706,7 +706,8 @@ Check if they're satisfied or want further refinement. Then commit and push memo
 cd $MEMORY_DIR
 git status                # Review what changed before staging
 git add <specific files>  # Stage targeted paths — avoid blind `git add -A`
-git commit --author="<AGENT_NAME> <<ACTUAL_AGENT_ID>@letta.com>" -m "feat(init): <summary> ✨
+author_name="${AGENT_NAME:-$AGENT_ID}"
+git commit --author="$author_name <$AGENT_ID@letta.com>" -m "feat(init): <summary> ✨
 
 <what was initialized and key decisions made>"