jeo-code 0.8.39 → 0.8.40

package/CHANGELOG.md

@@ -6,6 +6,23 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 The README mirrors the latest 5 entries — regenerate with `bun run changelog:sync`.
 
+## [0.8.40] - 2026-07-21
+_User report (paraphrased): "the /route command doesn't work, fix it." Investigated live before assuming a defect — booted a real `jeo --tmux` session with real Anthropic OAuth credentials from the actual `~/.jeo/config.json` and drove `/route status|on|off|why|history|bogus` plus real end-to-end routed turns. Every subcommand and the routing engine itself (heuristic tiering, `roles.smol` resolution, live provider call, per-turn "thinking" card showing the routed model) worked correctly, including a real stale-catalog edge case (`roles.smol: "claude-haiku-4-5"`, removed from jeo's own catalog in an earlier curation pass, still being accepted directly by the Anthropic API). The one genuine, reproducible gap: `/route on` is SESSION-LOCAL ONLY — `config.routing.enabled` in `~/.jeo/config.json` is never written, so a user has to retype `/route on` at the start of every single session, and (separately, working as designed but easy to miss) an explicit `--model`/`/model` pin silently blocks routing from ever engaging unless `/route on` is run to override it that session. The "settings don't stick" experience is the most likely reading of "doesn't work."_
+
+### Added
+- **`/route on save` / `/route off save` / bare `/route save`** (`src/commands/launch/route-slash.ts`) — persists `routing.enabled` to `~/.jeo/config.json` (mirrors `/model save`'s existing session-vs-persistent split), so routing preference now survives a restart without re-toggling every session. The bare form persists whatever is currently effective (an active session override, else the config's existing value); the `on`/`off save` forms also set the session-local override in the same call. All three carefully merge into the existing `routing` object rather than replacing it, so previously configured `tiers`/`confidenceThreshold`/`crossProviderPool` are never clobbered.
+
+### Verified
+- Live (unmocked), with real credentials and a real config copy: `/route on save` wrote `{"routing":{"enabled":true}}` to disk; a fresh `jeo --tmux` process launched against that same config dir immediately reported `routing: on (this session)` via `/route status` with zero re-toggling.
+- `test/route-slash.test.ts` — 28 pass (was 22): 6 new tests cover `on save`/`off save`/bare `save`, config-field preservation (tiers/confidenceThreshold untouched), and confirming a plain `/route on` (no `save`) never touches disk.
+- Full `bun test` — 3130 pass / 4 pre-existing unrelated flakes (browser screenshot/close timing, LSP definition timing, a perf-threshold test) across 307 files, all 4 confirmed passing in isolated re-runs.
+- `bun run typecheck` — no errors.
+
+### Fixed
+- **Telegram daemon lock PID-reuse safety** (`src/agent/notify/daemon-control.ts`, gjc #2786 parity: "restore macOS daemon signaling (kill(2) + start-time recheck)") — `isPidAlive`'s bare `kill(pid, 0)` cannot distinguish a live daemon from an unrelated process that later reused the same pid after the daemon crashed; reproduced live (`isPidAlive(process.pid)` on an arbitrary unrelated process reports "alive"). New `isLockOwnerAlive` cross-checks the recorded `startedAt` against the OS-reported real process start time (`ps -o etime=`, parsed and diffed with a 10s tolerance) before trusting a lock's pid — used by `acquireDaemonLock`, `daemonStatus`, `startDaemon`, and (most importantly) `stopDaemon`, which previously could have sent a real `SIGTERM` to a completely unrelated process. Falls back to the old existence-only check whenever the OS lookup itself is unavailable (Windows, `ps` missing) — never a regression there.
+- `test/notify-daemon-control.test.ts` — 22 pass (was 19): 3 new tests directly reproduce PID reuse (mismatched `startedAt` on a live pid reports not-alive; `stopDaemon` refuses to signal the mismatched pid and clears the lock instead, verified the unrelated live process survives).
+- `test/daemon-command.test.ts` — fixed to use a real process start time instead of an arbitrary placeholder timestamp (the new stricter check correctly flagged the old fixture as a mismatch).
+
 ## [0.8.39] - 2026-07-21
 _User report (paraphrased): "when I copy and paste content from elsewhere, it doesn't need to show every line — like gjc — update it so the composer stays a reasonable size and shows a clear indicator when width/length isn't enough." Investigated live before assuming intent: pasting a large multi-line block into the composer already caps the visible box height and scrolls to the caret, but the scrolled-off content above/below was flagged with a bare `…` — no indication of HOW MUCH text was hidden. (Checked gjc's own actual paste-handling source directly: it does not have a general text-paste-collapse feature either — only image-path detection — so this fix targets the concrete, reproducible UX gap found live, not a specific gjc feature to port.)_
 

package/README.ja.md

@@ -283,11 +283,11 @@ CI は `.github/workflows/npm-publish.yml` で公開します — GitHub リリ
 ## 変更履歴 (Changelog)
 
 <!-- CHANGELOG:START (auto-generated from CHANGELOG.md — run `bun run changelog:sync`) -->
+- **[0.8.40]** (2026-07-21) — User report (paraphrased): "the /route command doesn't work, fix it." Investigated live before assuming a defect — booted a real `jeo --tmux` session with real Anthropic OAuth credentials from the actual `~/.jeo/config.json` and drove `/route status|on|off|why|history|bogus` plus real end-to-end routed turns. Every subcommand and the routing engine itself (heuristic tiering, `roles.smol` resolution, live provider call, per-turn "thinking" card showing the routed model) worked correctly, including a real stale-catalog edge case (`roles.smol: "claude-haiku-4-5"`, removed from jeo's own catalog in an earlier curation pass, still being accepted directly by the Anthropic API). The one genuine, reproducible gap: `/route on` is SESSION-LOCAL ONLY — `config.routing.enabled` in `~/.jeo/config.json` is never written, so a user has to retype `/route on` at the start of every single session, and (separately, working as designed but easy to miss) an explicit `--model`/`/model` pin silently blocks routing from ever engaging unless `/route on` is run to override it that session. The "settings don't stick" experience is the most likely reading of "doesn't work."
 - **[0.8.39]** (2026-07-21) — User report (paraphrased): "when I copy and paste content from elsewhere, it doesn't need to show every line — like gjc — update it so the composer stays a reasonable size and shows a clear indicator when width/length isn't enough." Investigated live before assuming intent: pasting a large multi-line block into the composer already caps the visible box height and scrolls to the caret, but the scrolled-off content above/below was flagged with a bare `…` — no indication of HOW MUCH text was hidden. (Checked gjc's own actual paste-handling source directly: it does not have a general text-paste-collapse feature either — only image-path detection — so this fix targets the concrete, reproducible UX gap found live, not a specific gjc feature to port.)
 - **[0.8.38]** (2026-07-21) — Independent architect-subagent review of this whole session's TUI/workflow changes (v0.8.29-v0.8.37), requested explicitly to verify stability before calling the work done. Verdict: WATCH/COMMENT, no P0/P1 — the resize fix, the idle↔mid-turn transition, the new `calc` tool wiring, and the web-search credential gating all held up under independent scrutiny. Found and fixed 3 real, previously-unknown bugs in `width.ts`; 1 architectural observation recorded as tech debt, not fixed.
 - **[0.8.37]** (2026-07-21) — Long-term goal opened to triage every remaining PR in gjc v0.11.5's ~65-entry list, one batch at a time. This entry covers the first batch (10 PRs read in full).
 - **[0.8.36]** (2026-07-21) — First full audit of gjc v0.11.5's actual ~65-PR "What's Changed" list (prior passes had used release-note prose or direct-source structural diffs — this is the first time every individual PR was read and triaged against jeo's own codebase).
-- **[0.8.35]** (2026-07-21) — Deeper gjc-parity pass: re-cloned the real, public `Yeachan-Heo/gajae-code` source and, this time, structurally compared its actual tool implementations (`packages/coding-agent/src/tools/`) against jeo's own tool surface, not just the provider catalog.
 
 See [CHANGELOG.md](CHANGELOG.md) for the full history.
 <!-- CHANGELOG:END -->

package/README.ko.md

@@ -284,11 +284,11 @@ CI는 `.github/workflows/npm-publish.yml`로 배포합니다 — GitHub 릴리
 ## 변경 이력 (Changelog)
 
 <!-- CHANGELOG:START (auto-generated from CHANGELOG.md — run `bun run changelog:sync`) -->
+- **[0.8.40]** (2026-07-21) — User report (paraphrased): "the /route command doesn't work, fix it." Investigated live before assuming a defect — booted a real `jeo --tmux` session with real Anthropic OAuth credentials from the actual `~/.jeo/config.json` and drove `/route status|on|off|why|history|bogus` plus real end-to-end routed turns. Every subcommand and the routing engine itself (heuristic tiering, `roles.smol` resolution, live provider call, per-turn "thinking" card showing the routed model) worked correctly, including a real stale-catalog edge case (`roles.smol: "claude-haiku-4-5"`, removed from jeo's own catalog in an earlier curation pass, still being accepted directly by the Anthropic API). The one genuine, reproducible gap: `/route on` is SESSION-LOCAL ONLY — `config.routing.enabled` in `~/.jeo/config.json` is never written, so a user has to retype `/route on` at the start of every single session, and (separately, working as designed but easy to miss) an explicit `--model`/`/model` pin silently blocks routing from ever engaging unless `/route on` is run to override it that session. The "settings don't stick" experience is the most likely reading of "doesn't work."
 - **[0.8.39]** (2026-07-21) — User report (paraphrased): "when I copy and paste content from elsewhere, it doesn't need to show every line — like gjc — update it so the composer stays a reasonable size and shows a clear indicator when width/length isn't enough." Investigated live before assuming intent: pasting a large multi-line block into the composer already caps the visible box height and scrolls to the caret, but the scrolled-off content above/below was flagged with a bare `…` — no indication of HOW MUCH text was hidden. (Checked gjc's own actual paste-handling source directly: it does not have a general text-paste-collapse feature either — only image-path detection — so this fix targets the concrete, reproducible UX gap found live, not a specific gjc feature to port.)
 - **[0.8.38]** (2026-07-21) — Independent architect-subagent review of this whole session's TUI/workflow changes (v0.8.29-v0.8.37), requested explicitly to verify stability before calling the work done. Verdict: WATCH/COMMENT, no P0/P1 — the resize fix, the idle↔mid-turn transition, the new `calc` tool wiring, and the web-search credential gating all held up under independent scrutiny. Found and fixed 3 real, previously-unknown bugs in `width.ts`; 1 architectural observation recorded as tech debt, not fixed.
 - **[0.8.37]** (2026-07-21) — Long-term goal opened to triage every remaining PR in gjc v0.11.5's ~65-entry list, one batch at a time. This entry covers the first batch (10 PRs read in full).
 - **[0.8.36]** (2026-07-21) — First full audit of gjc v0.11.5's actual ~65-PR "What's Changed" list (prior passes had used release-note prose or direct-source structural diffs — this is the first time every individual PR was read and triaged against jeo's own codebase).
-- **[0.8.35]** (2026-07-21) — Deeper gjc-parity pass: re-cloned the real, public `Yeachan-Heo/gajae-code` source and, this time, structurally compared its actual tool implementations (`packages/coding-agent/src/tools/`) against jeo's own tool surface, not just the provider catalog.
 
 See [CHANGELOG.md](CHANGELOG.md) for the full history.
 <!-- CHANGELOG:END -->

package/README.md

@@ -291,11 +291,11 @@ Huge thanks to [gajae-code](https://github.com/Yeachan-Heo/gajae-code) for the i
 ## Changelog
 
 <!-- CHANGELOG:START (auto-generated from CHANGELOG.md — run `bun run changelog:sync`) -->
+- **[0.8.40]** (2026-07-21) — User report (paraphrased): "the /route command doesn't work, fix it." Investigated live before assuming a defect — booted a real `jeo --tmux` session with real Anthropic OAuth credentials from the actual `~/.jeo/config.json` and drove `/route status|on|off|why|history|bogus` plus real end-to-end routed turns. Every subcommand and the routing engine itself (heuristic tiering, `roles.smol` resolution, live provider call, per-turn "thinking" card showing the routed model) worked correctly, including a real stale-catalog edge case (`roles.smol: "claude-haiku-4-5"`, removed from jeo's own catalog in an earlier curation pass, still being accepted directly by the Anthropic API). The one genuine, reproducible gap: `/route on` is SESSION-LOCAL ONLY — `config.routing.enabled` in `~/.jeo/config.json` is never written, so a user has to retype `/route on` at the start of every single session, and (separately, working as designed but easy to miss) an explicit `--model`/`/model` pin silently blocks routing from ever engaging unless `/route on` is run to override it that session. The "settings don't stick" experience is the most likely reading of "doesn't work."
 - **[0.8.39]** (2026-07-21) — User report (paraphrased): "when I copy and paste content from elsewhere, it doesn't need to show every line — like gjc — update it so the composer stays a reasonable size and shows a clear indicator when width/length isn't enough." Investigated live before assuming intent: pasting a large multi-line block into the composer already caps the visible box height and scrolls to the caret, but the scrolled-off content above/below was flagged with a bare `…` — no indication of HOW MUCH text was hidden. (Checked gjc's own actual paste-handling source directly: it does not have a general text-paste-collapse feature either — only image-path detection — so this fix targets the concrete, reproducible UX gap found live, not a specific gjc feature to port.)
 - **[0.8.38]** (2026-07-21) — Independent architect-subagent review of this whole session's TUI/workflow changes (v0.8.29-v0.8.37), requested explicitly to verify stability before calling the work done. Verdict: WATCH/COMMENT, no P0/P1 — the resize fix, the idle↔mid-turn transition, the new `calc` tool wiring, and the web-search credential gating all held up under independent scrutiny. Found and fixed 3 real, previously-unknown bugs in `width.ts`; 1 architectural observation recorded as tech debt, not fixed.
 - **[0.8.37]** (2026-07-21) — Long-term goal opened to triage every remaining PR in gjc v0.11.5's ~65-entry list, one batch at a time. This entry covers the first batch (10 PRs read in full).
 - **[0.8.36]** (2026-07-21) — First full audit of gjc v0.11.5's actual ~65-PR "What's Changed" list (prior passes had used release-note prose or direct-source structural diffs — this is the first time every individual PR was read and triaged against jeo's own codebase).
-- **[0.8.35]** (2026-07-21) — Deeper gjc-parity pass: re-cloned the real, public `Yeachan-Heo/gajae-code` source and, this time, structurally compared its actual tool implementations (`packages/coding-agent/src/tools/`) against jeo's own tool surface, not just the provider catalog.
 
 See [CHANGELOG.md](CHANGELOG.md) for the full history.
 <!-- CHANGELOG:END -->

package/README.zh.md

@@ -283,11 +283,11 @@ CI 通过 `.github/workflows/npm-publish.yml` 发布 — GitHub 发布 release
 ## 更新日志 (Changelog)
 
 <!-- CHANGELOG:START (auto-generated from CHANGELOG.md — run `bun run changelog:sync`) -->
+- **[0.8.40]** (2026-07-21) — User report (paraphrased): "the /route command doesn't work, fix it." Investigated live before assuming a defect — booted a real `jeo --tmux` session with real Anthropic OAuth credentials from the actual `~/.jeo/config.json` and drove `/route status|on|off|why|history|bogus` plus real end-to-end routed turns. Every subcommand and the routing engine itself (heuristic tiering, `roles.smol` resolution, live provider call, per-turn "thinking" card showing the routed model) worked correctly, including a real stale-catalog edge case (`roles.smol: "claude-haiku-4-5"`, removed from jeo's own catalog in an earlier curation pass, still being accepted directly by the Anthropic API). The one genuine, reproducible gap: `/route on` is SESSION-LOCAL ONLY — `config.routing.enabled` in `~/.jeo/config.json` is never written, so a user has to retype `/route on` at the start of every single session, and (separately, working as designed but easy to miss) an explicit `--model`/`/model` pin silently blocks routing from ever engaging unless `/route on` is run to override it that session. The "settings don't stick" experience is the most likely reading of "doesn't work."
 - **[0.8.39]** (2026-07-21) — User report (paraphrased): "when I copy and paste content from elsewhere, it doesn't need to show every line — like gjc — update it so the composer stays a reasonable size and shows a clear indicator when width/length isn't enough." Investigated live before assuming intent: pasting a large multi-line block into the composer already caps the visible box height and scrolls to the caret, but the scrolled-off content above/below was flagged with a bare `…` — no indication of HOW MUCH text was hidden. (Checked gjc's own actual paste-handling source directly: it does not have a general text-paste-collapse feature either — only image-path detection — so this fix targets the concrete, reproducible UX gap found live, not a specific gjc feature to port.)
 - **[0.8.38]** (2026-07-21) — Independent architect-subagent review of this whole session's TUI/workflow changes (v0.8.29-v0.8.37), requested explicitly to verify stability before calling the work done. Verdict: WATCH/COMMENT, no P0/P1 — the resize fix, the idle↔mid-turn transition, the new `calc` tool wiring, and the web-search credential gating all held up under independent scrutiny. Found and fixed 3 real, previously-unknown bugs in `width.ts`; 1 architectural observation recorded as tech debt, not fixed.
 - **[0.8.37]** (2026-07-21) — Long-term goal opened to triage every remaining PR in gjc v0.11.5's ~65-entry list, one batch at a time. This entry covers the first batch (10 PRs read in full).
 - **[0.8.36]** (2026-07-21) — First full audit of gjc v0.11.5's actual ~65-PR "What's Changed" list (prior passes had used release-note prose or direct-source structural diffs — this is the first time every individual PR was read and triaged against jeo's own codebase).
-- **[0.8.35]** (2026-07-21) — Deeper gjc-parity pass: re-cloned the real, public `Yeachan-Heo/gajae-code` source and, this time, structurally compared its actual tool implementations (`packages/coding-agent/src/tools/`) against jeo's own tool surface, not just the provider catalog.
 
 See [CHANGELOG.md](CHANGELOG.md) for the full history.
 <!-- CHANGELOG:END -->

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "jeo-code",
-  "version": "0.8.39",
+  "version": "0.8.40",
   "description": "Clean, highly optimized AI coding agent using spec-first loop",
   "type": "module",
   "main": "src/cli.ts",

package/src/agent/notify/daemon-control.ts

@@ -25,6 +25,67 @@ export function isPidAlive(pid: number): boolean {
     return false;
   }
 }
+export function parseEtimeToMs(raw: string): number | undefined {
+  const trimmed = raw.trim();
+  if (!trimmed) return undefined;
+  const dayMatch = trimmed.match(/^(\d+)-(.+)$/);
+  let days = 0;
+  let rest = trimmed;
+  if (dayMatch) {
+    days = Number(dayMatch[1]);
+    rest = dayMatch[2]!;
+  }
+  const parts = rest.split(":").map(Number);
+  if (parts.length < 2 || parts.length > 3 || parts.some(n => !Number.isFinite(n))) return undefined;
+  const [h, m, s] = parts.length === 3 ? parts : [0, parts[0]!, parts[1]!];
+  return ((days * 24 + h!) * 60 + m!) * 60 * 1000 + s! * 1000;
+}
+
+/** Best-effort actual OS process start time in epoch ms, via `ps -o etime=`
+ *  (POSIX-portable across macOS/Linux `ps`; unsupported on Windows, where
+ *  there is no equivalent zero-dependency lookup). Returns `undefined` when
+ *  the lookup is unavailable or unparseable — callers MUST treat that as
+ *  "cannot verify" and fall back to the existence-only check, not as "dead". */
+export async function processStartTimeMs(pid: number): Promise<number | undefined> {
+  if (process.platform === "win32") return undefined;
+  const output = await new Promise<string | undefined>(resolve => {
+    let child: ReturnType<typeof nodeSpawn>;
+    try {
+      child = nodeSpawn("ps", ["-o", "etime=", "-p", String(pid)], { stdio: ["ignore", "pipe", "ignore"] });
+    } catch {
+      resolve(undefined);
+      return;
+    }
+    let buf = "";
+    child.stdout?.on("data", d => { buf += d; });
+    child.on("error", () => resolve(undefined));
+    child.on("close", code => resolve(code === 0 ? buf : undefined));
+  });
+  if (output === undefined) return undefined;
+  const ms = parseEtimeToMs(output);
+  return ms === undefined ? undefined : Date.now() - ms;
+}
+
+/** `etime` is truncated to whole seconds and adds a little spawn/poll jitter;
+ *  genuine PID reuse shows up as a gap of minutes/hours/days, never a few
+ *  seconds, so a generous-but-tight tolerance loses no real detections. */
+const START_TIME_TOLERANCE_MS = 10_000;
+
+/** Stronger alternative to a bare `isPidAlive(lock.pid)`: also cross-checks
+ *  the recorded `startedAt` against the OS-reported actual process start
+ *  time to rule out PID reuse — a dead daemon's pid can be reassigned by the
+ *  OS to a completely unrelated process before the next status/stop check
+ *  runs, which `kill(pid, 0)` alone cannot distinguish (gjc #2786 parity:
+ *  "restore macOS daemon signaling (kill(2) + start-time recheck)"). Falls
+ *  back to the existence-only result whenever the OS lookup itself is
+ *  unavailable (Windows, `ps` missing, permission denied) — never a
+ *  regression versus the old behavior in that case. */
+export async function isLockOwnerAlive(lock: DaemonLockInfo): Promise<boolean> {
+  if (!isPidAlive(lock.pid)) return false;
+  const real = await processStartTimeMs(lock.pid);
+  if (real === undefined) return true;
+  return Math.abs(real - lock.startedAt) <= START_TIME_TOLERANCE_MS;
+}
 
 export async function readDaemonLock(): Promise<DaemonLockInfo | undefined> {
   try {
@@ -43,7 +104,7 @@ export async function readDaemonLock(): Promise<DaemonLockInfo | undefined> {
 export async function acquireDaemonLock(): Promise<{ release: () => Promise<void> } | undefined> {
   await fs.mkdir(notifyDir(), { recursive: true, mode: 0o700 });
   const existing = await readDaemonLock();
-  if (existing && existing.pid !== process.pid && isPidAlive(existing.pid)) return undefined;
+  if (existing && existing.pid !== process.pid && (await isLockOwnerAlive(existing))) return undefined;
   const info: DaemonLockInfo = { pid: process.pid, startedAt: Date.now() };
   await fs.writeFile(notifyDaemonLockPath(), JSON.stringify(info), { mode: 0o600 });
   return {
@@ -71,7 +132,7 @@ export async function isNotifyConfigured(): Promise<boolean> {
 export async function daemonStatus(): Promise<DaemonStatus> {
   const [lock, configured] = await Promise.all([readDaemonLock(), isNotifyConfigured()]);
   if (!lock) return { configured, running: false, stale: false };
-  const alive = isPidAlive(lock.pid);
+  const alive = await isLockOwnerAlive(lock);
   return { configured, running: alive, stale: !alive, pid: lock.pid, startedAt: lock.startedAt };
 }
 
@@ -115,7 +176,7 @@ async function waitFor(predicate: () => Promise<boolean>, timeoutMs: number, ste
 
 export async function startDaemon(spawnImpl: SpawnLike = defaultSpawn): Promise<{ ok: boolean; pid?: number; message: string }> {
   const existing = await readDaemonLock();
-  if (existing && isPidAlive(existing.pid)) {
+  if (existing && (await isLockOwnerAlive(existing))) {
     return { ok: true, pid: existing.pid, message: `daemon already running (pid ${existing.pid})` };
   }
   // Check BEFORE spawning: an unconfigured daemon exits almost immediately (see
@@ -138,7 +199,7 @@ export async function startDaemon(spawnImpl: SpawnLike = defaultSpawn): Promise<
 
 export async function stopDaemon(): Promise<{ ok: boolean; message: string }> {
   const lock = await readDaemonLock();
-  if (!lock || !isPidAlive(lock.pid)) {
+  if (!lock || !(await isLockOwnerAlive(lock))) {
     await fs.unlink(notifyDaemonLockPath()).catch(() => {});
     return { ok: true, message: "daemon was not running" };
   }

package/src/commands/launch/route-slash.ts

@@ -1,8 +1,8 @@
 /**
  * `/route` slash-command handler extracted from launch.ts.
  *
- * Handles `/route [status|on|off|why|history]` (default/`status`: report the
- * effective session routing state + the last decision, if any; `on`/`off`:
+ * Handles `/route [status|on|off|why|history|save]` (default/`status`: report
+ * the effective session routing state + the last decision, if any; `on`/`off`:
  * toggle a SESSION-LOCAL override of `config.routing.enabled`; `why`: explain
  * the last routing decision in detail; `history [n]`: print the last `n`
  * (default 10) recorded decisions, one per line). This block shares mutable
@@ -11,14 +11,16 @@
  * values in via an explicit context object and reads back a result object
  * instead of this function closing over REPL state directly.
  *
- * Session-level toggle only — `sessionRouteOverride` is never persisted to
- * `~/.jeo/config.json` (mirrors `/thinking` being session-only vs. `/model
- * save` being the explicit opt-in for persistence). `/route` has no `save`
- * subcommand in v1.
+ * `on`/`off` alone are SESSION-LOCAL only (mirrors `/thinking`), same as
+ * before — but `on save`/`off save`/bare `save` now persist `routing.enabled`
+ * to `~/.jeo/config.json` (mirrors `/model save`'s explicit opt-in for
+ * persistence), so a user who wants routing on by default no longer has to
+ * remember to type `/route on` at the start of every session.
  */
 
 import type { RouteDecision } from "../../agent/prompt-router";
 import type { RouteHistoryEntry } from "../../agent/route-history";
+import { saveConfigPatch } from "../../agent/state";
 
 export interface RouteSlashCtx {
   sessionRouteOverride: boolean | undefined;
@@ -44,7 +46,7 @@ export interface RouteSlashResult {
   lines: string[];
 }
 
-const USAGE = "Usage: /route [status|on|off|why|history [n]]";
+const USAGE = "Usage: /route [status|on|off|why|history [n]|save|on save|off save]";
 
 function isRealDecision(decision: RouteDecision | { note: string } | null): decision is RouteDecision {
   return !!decision && "model" in decision;
@@ -82,11 +84,11 @@ function formatHistory(entries: RouteHistoryEntry[], n: number): string[] {
 }
 
 /**
- * Handle `/route [status|on|off|why|history]`. Extracted for the same reason
- * as `/model`: shares REPL-local routing state with `runTurn` via an
+ * Handle `/route [status|on|off|why|history|save]`. Extracted for the same
+ * reason as `/model`: shares REPL-local routing state with `runTurn` via an
  * explicit ctx/result object rather than closing over it.
  */
-export function runRouteSlash(input: string, ctx: RouteSlashCtx): RouteSlashResult {
+export async function runRouteSlash(input: string, ctx: RouteSlashCtx): Promise<RouteSlashResult> {
   const rest = input.slice("/route".length).trim();
   const [sub, arg] = rest.split(/\s+/).filter(Boolean);
   const effective = ctx.sessionRouteOverride ?? ctx.routingConfigEnabled;
@@ -111,12 +113,21 @@ export function runRouteSlash(input: string, ctx: RouteSlashCtx): RouteSlashResu
 
 
 
-  if (sub === "on") {
-    return { sessionRouteOverride: true, lines: ["routing: on (this session)"] };
+  if (sub === "on" || sub === "off") {
+    const enabled = sub === "on";
+    if (arg === "save") {
+      await saveConfigPatch(raw => ({ routing: { ...raw.routing, enabled } }));
+      return { sessionRouteOverride: enabled, lines: [`routing: ${enabled ? "on" : "off"} (this session) — saved to ~/.jeo/config.json`] };
+    }
+    return { sessionRouteOverride: enabled, lines: [`routing: ${enabled ? "on" : "off"} (this session)`] };
   }
 
-  if (sub === "off") {
-    return { sessionRouteOverride: false, lines: ["routing: off (this session)"] };
+  // Bare `/route save` persists whatever is CURRENTLY effective (a prior
+  // session-local `on`/`off`, or — if neither was toggled this session —
+  // the config's existing value) rather than requiring `/route on save`.
+  if (sub === "save") {
+    await saveConfigPatch(raw => ({ routing: { ...raw.routing, enabled: effective } }));
+    return { lines: [`routing: ${effective ? "on" : "off"} — saved to ~/.jeo/config.json`] };
   }
 
   if (sub === "why") {

package/src/commands/launch.ts

@@ -4810,7 +4810,7 @@ export async function runLaunchCommand(args: string[]): Promise<void> {
       }
       if (input.startsWith("/route") && (input === "/route" || input[6] === " ")) {
         const cfgNow = await readGlobalConfig();
-        const routeResult = runRouteSlash(input, {
+        const routeResult = await runRouteSlash(input, {
           sessionRouteOverride,
           routingConfigEnabled: !!cfgNow.routing?.enabled,
           lastRouteDecision,

package/src/tui/components/slash.ts

@@ -43,7 +43,7 @@ export const SLASH_COMMAND_DETAILS: readonly SlashCommandInfo[] = [
   { command: "/logout", usage: "/logout <anthropic|openai|gemini|antigravity>", description: "Remove the stored OAuth token for a provider", group: "models" },
   { command: "/roles", usage: "/roles [tier model]", description: "Show or set model role tiers (smol/slow/plan)", group: "models" },
   { command: "/thinking", usage: "/thinking [level]", description: "Show or set thinking budget (low/medium/high/xhigh)", group: "models" },
-  { command: "/route", usage: "/route [on|off|why|history [n]]", description: "Show/toggle prompt-based model routing for this session, explain the last routing decision, or list recent routing history", group: "models" },
+  { command: "/route", usage: "/route [on|off|why|history [n]|save|on save|off save]", description: "Show/toggle prompt-based model routing for this session, explain the last routing decision, or list recent routing history (add save to persist on/off)", group: "models" },
   { command: "/agents", usage: "/agents [edit|role] [model|#N|thinking L|maxSteps N|reset]", description: "List subagent roles; use /agents edit for the interactive picker or pin role model/settings", group: "subagents" },
   { command: "/subagent", usage: "/subagent [role]", description: "Show the current subagent composition (per-role model · thinking · steps); alias of /agents", group: "subagents" },
   { command: "/view", usage: "/view <file> [a-b]", description: "Render a file with line numbers + light highlight", group: "code" },