@letta-ai/letta-code 0.30.32 → 0.31.0

package/scripts/pi-ai-watch/tracker.test.ts

@@ -1,7 +1,6 @@
 import { describe, expect, test } from "bun:test";
 import type { PiAiWatchAnalysis } from "./release-analysis.ts";
 import {
-  advanceMergedPr,
   getPendingPrForCursor,
   hasCompletedRange,
   hasRecordedOutcome,
@@ -19,17 +18,18 @@ describe("pi-ai watch tracker", () => {
 
   test("advances after a no-upgrade decision", () => {
     const state = recordAnalysis(initialTrackerState("0.82.1"), {
-      analysis: analysis("0.82.1", "0.83.0"),
+      analysis: analysis("0.82.1", "0.84.0"),
       outcome: "no_upgrade",
       notes: "upstream-only provider change",
       processedAt: "2026-08-21T00:00:00Z",
     });
-    expect(state.audit_cursor_version).toBe("0.83.0");
-    expect(hasCompletedRange(state, "0.82.1", "0.83.0")).toBe(true);
-    expect(hasRecordedOutcome(state, "0.82.1", "0.83.0")).toBe(true);
+    expect(state.audit_cursor_version).toBe("0.84.0");
+    expect(hasCompletedRange(state, "0.82.1", "0.84.0")).toBe(true);
+    expect(hasRecordedOutcome(state, "0.82.1", "0.84.0")).toBe(true);
+    expect(renderTrackerBody(state)).toContain("[0.82.1...0.84.0]");
   });
 
-  test("keeps a created PR pending until it merges", () => {
+  test("advances a created PR and keeps it pending at the cursor", () => {
     const pending = recordAnalysis(initialTrackerState("0.82.1"), {
       analysis: analysis("0.82.1", "0.83.0"),
       outcome: "pr_created",
@@ -37,14 +37,18 @@ describe("pi-ai watch tracker", () => {
       prUrl: "https://github.com/letta-ai/letta-code/pull/123",
       processedAt: "2026-08-21T00:00:00Z",
     });
-    expect(pending.audit_cursor_version).toBe("0.82.1");
+    expect(pending.audit_cursor_version).toBe("0.83.0");
     expect(getPendingPrForCursor(pending)?.version).toBe("0.83.0");
-    expect(hasCompletedRange(pending, "0.82.1", "0.83.0")).toBe(false);
+    expect(hasCompletedRange(pending, "0.82.1", "0.83.0")).toBe(true);
     expect(hasRecordedOutcome(pending, "0.82.1", "0.83.0")).toBe(true);
 
-    const merged = advanceMergedPr(pending, "0.83.0");
-    expect(merged.audit_cursor_version).toBe("0.83.0");
-    expect(getPendingPrForCursor(merged)).toBeNull();
+    const later = recordAnalysis(pending, {
+      analysis: analysis("0.83.0", "0.84.0"),
+      outcome: "no_upgrade",
+      notes: "later release range",
+    });
+    expect(later.audit_cursor_version).toBe("0.84.0");
+    expect(getPendingPrForCursor(later)).toBeNull();
   });
 
   test("keeps errors retryable and replaces them with a later outcome", () => {
@@ -68,16 +72,16 @@ describe("pi-ai watch tracker", () => {
     expect(retried.processed[0]?.outcome).toBe("no_upgrade");
   });
 
-  test("advances uncertain releases but not explicit non-adjacent replays", () => {
+  test("advances cumulative reviews but not replays behind the cursor", () => {
     const human = recordAnalysis(initialTrackerState("0.82.1"), {
-      analysis: analysis("0.82.1", "0.83.0"),
+      analysis: analysis("0.82.1", "0.84.0"),
       outcome: "needs_human_review",
       notes: "product decision",
     });
-    expect(human.audit_cursor_version).toBe("0.83.0");
+    expect(human.audit_cursor_version).toBe("0.84.0");
 
     const replay = recordAnalysis(initialTrackerState("0.82.1"), {
-      analysis: analysis("0.82.1", "0.84.0", false),
+      analysis: analysis("0.82.1", "0.83.0", false),
       outcome: "no_upgrade",
       notes: "explicit replay",
     });
@@ -101,7 +105,7 @@ describe("pi-ai watch tracker", () => {
       const previous = `0.${index}.0`;
       const current = `0.${index}.1`;
       state = recordAnalysis(state, {
-        analysis: analysis(previous, current, false),
+        analysis: analysis(previous, current),
         outcome: "error",
         notes: `attempt ${index}`,
         processedAt: `2026-08-21T00:${String(index).padStart(2, "0")}:00Z`,
@@ -115,18 +119,21 @@ describe("pi-ai watch tracker", () => {
 function analysis(
   previousVersion: string,
   currentVersion: string,
-  adjacent = true,
+  isLatestRelease = true,
 ): PiAiWatchAnalysis {
   return {
     package: "@earendil-works/pi-ai",
     installed_version: "0.82.1",
     previous_version: previousVersion,
     current_version: currentVersion,
-    is_adjacent_release: adjacent,
+    is_latest_release: isLatestRelease,
     published_at: "2026-08-14T00:00:00Z",
     integrity: "sha512-test",
     tarball_url: "https://example.test/pi-ai.tgz",
+    previous_git_head: "previous123",
     git_head: "abc123",
+    previous_tag_commit: "previous123",
+    current_tag_commit: "abc123",
     release_url: `https://github.com/earendil-works/pi/releases/tag/v${currentVersion}`,
     compare_url: `https://github.com/earendil-works/pi/compare/v${previousVersion}...v${currentVersion}`,
     changelog_md: "## Added",

package/scripts/pi-ai-watch/tracker.ts

@@ -87,9 +87,9 @@ export function recordAnalysis(
   ].slice(0, HISTORY_LIMIT);
 
   const advancesCursor =
-    options.analysis.is_adjacent_release &&
+    options.analysis.is_latest_release &&
     entry.previous_version === state.audit_cursor_version &&
-    (entry.outcome === "no_upgrade" || entry.outcome === "needs_human_review");
+    entry.outcome !== "error";
 
   return {
     audit_cursor_version: advancesCursor
@@ -123,8 +123,7 @@ export function hasCompletedRange(
     (entry) =>
       entry.previous_version === previousVersion &&
       entry.version === currentVersion &&
-      (entry.outcome === "no_upgrade" ||
-        entry.outcome === "needs_human_review"),
+      entry.outcome !== "error",
   );
 }
 
@@ -134,24 +133,13 @@ export function getPendingPrForCursor(
   return (
     state.processed.find(
       (entry) =>
-        entry.previous_version === state.audit_cursor_version &&
+        entry.version === state.audit_cursor_version &&
         entry.outcome === "pr_created" &&
         entry.pr_url,
     ) ?? null
   );
 }
 
-export function advanceMergedPr(
-  state: TrackerState,
-  currentVersion: string,
-): TrackerState {
-  const pending = getPendingPrForCursor(state);
-  if (!pending || pending.version !== currentVersion) {
-    throw new Error(`No pending pi-ai PR for ${currentVersion}`);
-  }
-  return { ...state, audit_cursor_version: currentVersion };
-}
-
 export function renderTrackerBody(state: TrackerState): string {
   const normalized = normalizeState(state);
   return `${[
@@ -166,7 +154,7 @@ export function renderTrackerBody(state: TrackerState): string {
     "",
     "## Hidden state",
     "",
-    "The workflow uses the hidden JSON block below for ordered release processing and dedupe.",
+    "The workflow uses the hidden JSON block below for cumulative release processing and dedupe.",
     "",
     serializeTrackerState(normalized),
   ].join("\n")}\n`;
@@ -193,12 +181,12 @@ function renderTable(state: TrackerState): string {
   if (entries.length === 0) return "_No pi-ai releases reviewed yet._";
 
   const rows = [
-    "| Release | Installed | Outcome | PR | Notes |",
+    "| Range | Installed | Outcome | PR | Notes |",
     "|---|---|---|---|---|",
   ];
   for (const entry of entries) {
     rows.push(
-      `| [${entry.version}](${entry.compare_url}) | ${entry.installed_version} | ${entry.outcome} | ${renderPr(entry.pr_url)} | ${escapeTable(entry.notes)} |`,
+      `| [${entry.previous_version}...${entry.version}](${entry.compare_url}) | ${entry.installed_version} | ${entry.outcome} | ${renderPr(entry.pr_url)} | ${escapeTable(entry.notes)} |`,
     );
   }
   return rows.join("\n");

package/scripts/pi-ai-watch/update-tracker.ts

@@ -202,13 +202,13 @@ function parseOutcome(value: string | undefined): TrackerOutcome {
 function defaultNotes(outcome: TrackerOutcome): string {
   switch (outcome) {
     case "no_upgrade":
-      return "reviewed; no upgrade needed for this release";
+      return "reviewed; no upgrade needed for this release range";
     case "pr_created":
       return "opened pi-ai dependency upgrade PR";
     case "needs_human_review":
       return "needs human review";
     case "error":
-      return "automation hit an error; retry this release";
+      return "automation hit an error; retry this release range";
   }
 }
 

package/scripts/source-file-size-baseline.json

@@ -37,13 +37,13 @@
   "src/settings-manager.ts": 2102,
   "src/tools/manager.ts": 2993,
   "src/tools/tool-execution-context.test.ts": 1138,
-  "src/types/protocol_v2.ts": 2834,
+  "src/types/protocol_v2.ts": 2666,
   "src/websocket/listen-client-concurrency.test.ts": 2686,
   "src/websocket/listen-client-protocol.test.ts": 5819,
   "src/websocket/listener/commands/memory.ts": 1114,
   "src/websocket/listener/file-commands.ts": 1053,
   "src/websocket/listener/lifecycle.ts": 1048,
-  "src/websocket/listener/protocol-inbound.ts": 2237,
+  "src/websocket/listener/protocol-inbound.ts": 2227,
   "src/websocket/listener/protocol-outbound.ts": 1047,
   "src/websocket/listener/turn.ts": 1058
 }

package/skills/dispatching-coding-agents/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: dispatching-coding-agents
-description: Dispatch stateless coding agents (Claude Code or Codex) via Bash. Use when you're stuck, need a second opinion, or need parallel research on a hard problem. They have no memory — you must provide all context.
+description: Dispatch stateless coding agents through Claude Code or Codex. Model catalog checked 2026-08-25: Claude Code includes Sonnet 5, Opus 5, and Fable 5; Codex includes GPT-5.6 Luna, Terra, and Sol. Use when you're stuck, need a second opinion, or need parallel research on a hard problem. They have no memory, so you must provide all context.
 ---
 
 # Dispatching Coding Agents
@@ -43,14 +43,23 @@ Different agents have different strengths. Track what works in your memory over
 
 ### Categories
 
-**Codex:**
-- Use the configured default model. Model catalogs and account access change frequently, so only pass `--model` when the user explicitly requests one.
+**Codex (catalog checked 2026-08-25):**
+- `gpt-5.6-luna`: high-throughput, lower-latency work
+- `gpt-5.6-terra`: strong performance with a balance of intelligence and cost
+- `gpt-5.6-sol`: quality-first flagship for difficult coding and reasoning
+- Use the configured default unless the user requests a model. Pass a requested model with `-m`, for example `codex exec "YOUR PROMPT" -m gpt-5.6-sol --sandbox workspace-write`.
 - If a requested model is rejected, inspect the installed CLI and account configuration rather than guessing another model name.
 
-**Claude Code:**
-- `opus` — Excellent writer. Best for docs, refactors, open-ended tasks, and vague instructions.
-  - Strengths: Excellent writer, understands vague instructions, excellent for coding but also general-purpose
-  - Weaknesses: Tends to generate "slop", writing excessive quantities of code unnecessarily. Can hang on large repos.
+**Claude Code (catalog checked 2026-08-25):**
+- The current flagship models are Sonnet 5, Opus 5, and Fable 5. Use the configured default unless the user requests a model.
+- Claude Code's stable aliases follow its current catalog:
+  - `best`: highest-capability model available to the account, using Fable 5 when available
+  - `fable`: Fable 5 for the hardest, longest-running tasks
+  - `opus`: Opus 5 for complex reasoning, docs, refactors, and open-ended tasks
+  - `sonnet`: Sonnet 5 for everyday coding
+  - `haiku`: fast, simple tasks
+- Pass an alias with `--model`, for example `claude -p "YOUR PROMPT" --model fable --dangerously-skip-permissions`.
+- Aliases update as Claude Code's catalog changes. If an alias is rejected, inspect the installed CLI version and account access rather than guessing a versioned model name.
 
 ### Cost and speed tradeoffs
 - Use each CLI's configured default model unless the task requires a model the user explicitly requested