@lumoai/cli 1.56.0 → 1.57.0

package/assets/skill/SKILL.md

@@ -158,7 +158,7 @@ The command catalog below is a **map**: it lists every command grouped by domain
 
 **Verification (machine acceptance loop)** — see [verify.md](references/verify.md)
 
-- `lumo verify [task] --note "<self-report>" [--timeout <seconds>]` — run every MACHINE criterion's checkpointer locally and report a structured PASS/FAIL verdict per criterion. **`--note` is required** when a round will be posted: a one-line self-report ("我改了 X,因为 Y 可验收") frozen as your task's claim (source AGENT) the moment the round passes into IN_REVIEW, then checked against the diff for faithfulness. No note → no round posted (none burned). Round cap 3: all-pass → task to IN_REVIEW (agent stops); round-3 fail → escalate to human. **Run this before claiming a task is done.**
+- `lumo verify [task] --note "<self-report>" [--timeout <seconds>]` — run every MACHINE criterion's checkpointer locally and report a structured PASS/FAIL verdict per criterion. **`--note` is required** when a round will be posted: a one-line self-report ("我改了 X,因为 Y 可验收") frozen as your task's claim (source AGENT) the moment the round passes into IN_REVIEW, then checked against the diff for faithfulness. No note → no round posted (none burned). The round cap is the workspace's `verificationMaxRounds` (default 3) — read the actual budget off the `Round N/M` line the command prints, never assume a number. All-pass → task to IN_REVIEW (agent stops); a fail on the last round → escalate to human. **Run this before claiming a task is done.**
 - `lumo task status [task] [--json]` — read-only acceptance self-check (no LLM): the contract with each criterion's latest verdict, verification history/round, last failure reasons, `nextActions` (the unmet criteria), and any OPEN boundary crossings (fails closed — `null`/`⚠` means "could not confirm", not safe). **Run it first when resuming a task or after a round was rejected.**
 - `lumo verdict [task] --pass | --fail` — acceptance verdicts. `--pass` deep-links to the human verdict bar (records nothing). `--fail --reason <enum> …` records an AGENT send-back → IN_PROGRESS. **An unresolved send-back blocks the DONE transition with 409.**
 - `lumo crossing explain <id> --note "<text>"` — append an agent self-explanation ("申辩") to a boundary crossing; append-only, **never clears the crossing or unblocks Done** (disposition stays web + human-only). A review aid, not a self-clear.
@@ -302,7 +302,7 @@ Don't guess these:
 2. `lumo task context LUM-42` — load background; review unresolved items, PR-review todos, the description
 3. **If the task has no acceptance criteria** (context shows the draft reminder, not a contract): draft outcome-level criteria sized to the task (3–7 for a typical multi-file task; 1–2 for a micro task) and submit them with `lumo task criteria set` **before writing the first line of code** — see [criteria.md](references/criteria.md)
 4. Do the work
-5. **Before claiming done: `lumo verify`** — the machine half of the acceptance loop. Fix failures and re-run (round cap 3). All-pass → task moves to IN_REVIEW and you stop. See [verify.md](references/verify.md)
+5. **Before claiming done: `lumo verify`** — the machine half of the acceptance loop. Fix failures and re-run (the cap is workspace-configurable — see the `Round N/M` line in the output). All-pass → task moves to IN_REVIEW and you stop. See [verify.md](references/verify.md)
 
 **Resuming / after a send-back** — `lumo task status` **before** re-reading code or planning. It tells you where the loop stands (round, what passed, what's unmet and why, any REVIEW_ADDED criteria from review) so you don't redo finished work or miss why it bounced. A send-back means **fix in place / amend the contract — do not spin off a new task.** See [verify.md](references/verify.md)
 

package/assets/skill/references/criteria.md

@@ -219,4 +219,4 @@ lumo task criteria list LUM-42
 
 ## After the contract: the verification loop
 
-The contract is judged by `lumo verify` — run it before claiming the task is done. See [verify.md](verify.md) for the loop (round cap 3, IN_REVIEW on all-pass, escalation on a round-3 fail).
+The contract is judged by `lumo verify` — run it before claiming the task is done. See [verify.md](verify.md) for the loop (a workspace-configurable round cap — `verificationMaxRounds`, default 3, shown as `Round N/M` in the output — IN_REVIEW on all-pass, escalation on a last-round fail).

package/assets/skill/references/verify.md

@@ -3,8 +3,11 @@
 `lumo verify` is the machine half of the acceptance system: it executes every **MACHINE** criterion's checkpointer in the local
 repo, POSTs one structured PASS/FAIL verdict per criterion, and prints what to
 do next. Execution is on the client; adjudication is server-side — round
-numbering, the **3-round cap**, escalation, and the **IN_REVIEW** transition all
-happen on the server.
+numbering, the **round cap**, escalation, and the **IN_REVIEW** transition all
+happen on the server. The cap is **not a universal constant** — it is the
+workspace-level `verificationMaxRounds` (default 3); the effective value for
+_this_ workspace is the `M` in the `Round N/M` line that `lumo verify` and
+`lumo task status` print.
 
 ## The claim-done rule
 
@@ -39,11 +42,11 @@ still verifies — the claim degrades to the synthesized run-summary fallback.)
    TaskActivity event.
 4. Prints the round outcome:
 
-| Round outcome             | Effect                                                                                                                                                            | What to do                                                                                    |
-| ------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------- |
-| **All PASS**              | Task transitions to **IN_REVIEW** (existing state machine + TASK_IN_REVIEW notification); the `--note` self-report is frozen as the task's claim (source `AGENT`) | **Stop here.** Human adjudication + any HUMAN criteria take over; **never set DONE yourself** |
-| **Any FAIL**              | Task status untouched; unmet criteria printed as next actions (statement, checkpointer, failure tail)                                                             | Fix and re-run                                                                                |
-| **Round 3 still failing** | Loop escalates: a human is notified (AGENT_VERIFY, requires action); further `lumo verify` rounds are rejected with **409**                                       | **Stop retrying**; fix only what the human directs                                            |
+| Round outcome                                      | Effect                                                                                                                                                            | What to do                                                                                    |
+| -------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------- |
+| **All PASS**                                       | Task transitions to **IN_REVIEW** (existing state machine + TASK_IN_REVIEW notification); the `--note` self-report is frozen as the task's claim (source `AGENT`) | **Stop here.** Human adjudication + any HUMAN criteria take over; **never set DONE yourself** |
+| **Any FAIL**                                       | Task status untouched; unmet criteria printed as next actions (statement, checkpointer, failure tail)                                                             | Fix and re-run                                                                                |
+| **Last round still failing** (round = the cap `M`) | Loop escalates: a human is notified (AGENT_VERIFY, requires action); further `lumo verify` rounds are rejected with **409**                                       | **Stop retrying**; fix only what the human directs                                            |
 
 Exit code 0 = all passed (or nothing to run); 1 = failures, escalation, or
 errors.
@@ -74,9 +77,10 @@ errors.
 
 ## Round discipline
 
-Rounds are a hard budget of 3, not a retry loop. Between rounds, actually fix
-the failures — re-running without changes burns a round and (at round 3) pages a
-human. **A FAIL round never changes task status; only an all-pass round moves it
+Rounds are a hard budget, not a retry loop — and the size of that budget is a
+workspace setting (`verificationMaxRounds`, default 3), so check `Round N/M`
+instead of assuming three. Between rounds, actually fix the failures —
+re-running without changes burns a round and (on the last one) pages a human. **A FAIL round never changes task status; only an all-pass round moves it
 (to IN_REVIEW, never further).**
 
 ## Review-time drift habits (gap findings)
@@ -120,7 +124,7 @@ what's unmet and why (the exact failure tails), and how many rounds are left.
 
 ### What it prints
 
-- **Header** — task identifier/title/status + `verification round N/3` (round 0 = never verified) + an escalation warning when the machine loop is exhausted.
+- **Header** — task identifier/title/status + `verification round N/M` (`M` = the workspace's `verificationMaxRounds`; round 0 = never verified) + an escalation warning when the machine loop is exhausted.
 - **Claim vs verification** — 规律 2 声称vs核验: the headline contrast, printed right after the header (whenever the contract exists), so the report shows **both** columns instead of only the verification one. Two sides:
   - **Claim** — what the agent _says_ it did: an **unverified self-report** (`agent self-report · estimated, not verification`), estimate-tier provenance (估, not 测). Sourced by layering (LUM-597, preference `AGENT > RUN_SUMMARY > DIGEST > null`): if you supplied a `lumo verify --note`, that **own self-report** is the claim (`↳ source: agent self-report (verify --note)`) — the汇报者's voice, what faithfulness judges. Absent a self-report (old CLI / non-verify path) it degrades to the summarizer paraphrase, labelled honestly (`↳ source: synthesized run summary (no self-report)`). **Fail-closed**: when only a raw STOP turn digest exists it prints `generating — the formal run summary is still being synthesized` and **withholds the raw digest**; with no material at all it prints `not generated yet — …`, never a fabricated claim.
   - **Verification** — what was actually _confirmed_ (measured): the machine-verification rollup `N machine-verified / M human override (of T MACHINE criteria)` over the active MACHINE criteria (relocated here from its old standalone line under `Criteria`), plus `X of Y criteria met by their latest verdict`. **Fail-closed**: before any round runs it prints `no verification has run yet — the claim is unconfirmed` rather than implying a pass.
@@ -179,8 +183,9 @@ fields don't. Pin on `version` when scripting against it.
 - **`openCrossings` is `null` when the crossings check failed** — distinct from `[]` (a successful read with zero open crossings). Script consumers must treat `null` as "unknown / could not confirm", **not** "safe".
 
 `status` reads; `verify` judges. Running status never starts a round, never
-escalates, and never changes task state — loop rules (cap 3, IN_REVIEW on
-all-pass, human-only DONE) live entirely in `lumo verify` and the server.
+escalates, and never changes task state — loop rules (the workspace round cap,
+IN_REVIEW on all-pass, human-only DONE) live entirely in `lumo verify` and the
+server.
 
 ## lumo verdict — the three verdict channels
 

package/assets/skill/references/worktree.md

@@ -1,10 +1,24 @@
 # Worktrees
 
-Local dev tooling that scaffolds parallel git worktrees. Unlike every other
-`lumo` command, these do **not** touch the Lumo server — they shell out to git
-and the local filesystem. **Run `add` and `rm` from the main checkout**, not from
-inside a worktree (the node_modules symlink source and `.worktrees/` both live
-there); `list` is read-only and runnable from anywhere.
+Local dev tooling that scaffolds parallel git worktrees. These are almost
+entirely local — they shell out to git and the local filesystem; the **only**
+server call is a single `GET /api/teams` to expand a **bare number** to your
+default team prefix (see "Accepted identifier forms" below), and even that is
+skipped when you pass an explicit `<TEAM>-N`. **Run `add` and `rm` from the main
+checkout**, not from inside a worktree (the node_modules symlink source and
+`.worktrees/` both live there); `list` is read-only and runnable from anywhere.
+
+### Accepted identifier forms
+
+The `<id>` argument to `add`/`rm` is a task id with **any team prefix**, not
+just `LUM-` (the `<LUM-N>` in the examples below is just a placeholder):
+
+- `LUM-267`, `SPEC-12`, `spec-12` — an explicit `<TEAM>-<n>` (1–10 letters,
+  case-insensitive; the prefix is upper-cased). Resolved locally, no network.
+- `267` — a **bare number**: expanded to the workspace's **default team**
+  prefix via `GET /api/teams` (or the sole team in a single-team workspace).
+  In a multi-team workspace with no default, this errors and asks for an
+  explicit prefix; the same happens when you're offline / not logged in.
 
 ## `lumo worktree add <LUM-N> [slug]`
 

package/dist/cli/src/commands/worktree-add.js

@@ -39,6 +39,7 @@ const path = __importStar(require("path"));
 const child_process_1 = require("child_process");
 const next_steps_1 = require("../lib/next-steps");
 const worktree_1 = require("../lib/worktree");
+const worktree_ref_1 = require("../lib/worktree-ref");
 function printGuidance(dir, branch) {
     console.log('');
     console.log(`✓ Worktree ready: ${dir}`);
@@ -59,9 +60,16 @@ function printGuidance(dir, branch) {
     console.log('          cli/package.json haste collision silently runs the wrong tests.');
 }
 async function worktreeAdd(rawId, slug, opts) {
-    const taskId = (0, worktree_1.normalizeTaskId)(rawId);
+    let taskId;
+    try {
+        taskId = await (0, worktree_ref_1.resolveWorktreeTaskId)(rawId);
+    }
+    catch (err) {
+        console.error(`Error: ${err instanceof Error ? err.message : String(err)}`);
+        return 1;
+    }
     if (!taskId) {
-        console.error(`Error: "${rawId}" is not a task id (expected LUM-N or N)`);
+        console.error(`Error: "${rawId}" is not a task id (expected <TEAM>-N like LUM-42 or SPEC-12, or a bare number)`);
         return 1;
     }
     if (!(0, worktree_1.isMainCheckout)()) {

package/dist/cli/src/commands/worktree-list.js

@@ -46,7 +46,7 @@ function nodeModulesLinked(dir) {
         return false;
     }
 }
-/** Pull a LUM-N from a worktree dir basename or its branch, else '—'. */
+/** Pull a `<TEAM>-N` id from a worktree dir basename or its branch, else '—'. */
 function taskIdOf(dirBase, branch) {
     return ((0, git_task_1.matchTaskIdentifier)(dirBase) ??
         (branch ? (0, git_task_1.matchTaskIdentifier)(branch) : null) ??

package/dist/cli/src/commands/worktree-rm.js

@@ -2,10 +2,18 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.worktreeRm = worktreeRm;
 const worktree_1 = require("../lib/worktree");
+const worktree_ref_1 = require("../lib/worktree-ref");
 async function worktreeRm(rawId, opts) {
-    const taskId = (0, worktree_1.normalizeTaskId)(rawId);
+    let taskId;
+    try {
+        taskId = await (0, worktree_ref_1.resolveWorktreeTaskId)(rawId);
+    }
+    catch (err) {
+        console.error(`Error: ${err instanceof Error ? err.message : String(err)}`);
+        return 1;
+    }
     if (!taskId) {
-        console.error(`Error: "${rawId}" is not a task id (expected LUM-N or N)`);
+        console.error(`Error: "${rawId}" is not a task id (expected <TEAM>-N like LUM-42 or SPEC-12, or a bare number)`);
         return 1;
     }
     if (!(0, worktree_1.isMainCheckout)()) {

package/dist/cli/src/lib/worktree-ref.js

@@ -0,0 +1,53 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.resolveWorktreeTaskId = resolveWorktreeTaskId;
+const config_1 = require("./config");
+const api_1 = require("./api");
+const worktree_1 = require("./worktree");
+/**
+ * Resolve a user-supplied worktree ref to a canonical `<TEAM>-<n>` id.
+ *
+ * - Explicit prefix (`SPEC-12`, `lum-267`) → normalized locally, NO network.
+ * - Bare number (`267`) → completed with the workspace's DEFAULT team prefix
+ *   (GET /api/teams: the `isDefault` team, or the sole team in a single-team
+ *   workspace). Worktree tooling is otherwise purely local; the API is touched
+ *   only to disambiguate a prefix-less number.
+ *
+ * Returns null when `raw` is not a task-id shape at all (caller prints the
+ * "expected <TEAM>-N" usage error). Throws Error with a user-facing message
+ * when a bare number needs the default team but it can't be determined
+ * (not logged in / offline / ambiguous multi-team without a default) — every
+ * message points the user at passing an explicit prefix instead.
+ */
+async function resolveWorktreeTaskId(raw) {
+    const explicit = (0, worktree_1.normalizeTaskId)(raw);
+    if (explicit)
+        return explicit;
+    const bare = raw.trim().match(/^(\d+)$/);
+    if (!bare)
+        return null;
+    const n = bare[1];
+    const creds = (0, config_1.readCredentials)();
+    if (!creds) {
+        throw new Error(`a bare number needs your default team prefix — run \`lumo auth login\`, or pass an explicit id like LUM-${n}`);
+    }
+    const base = (0, api_1.trimTrailingSlash)((0, api_1.resolveAuthedApiUrl)(creds.apiUrl));
+    let teams;
+    try {
+        const res = await fetch(`${base}/api/teams`, {
+            headers: { Authorization: `Bearer ${creds.token}` },
+        });
+        if (!res.ok)
+            throw new Error(`HTTP ${res.status}`);
+        ({ teams } = (await res.json()));
+    }
+    catch (err) {
+        const msg = err instanceof Error ? err.message : String(err);
+        throw new Error(`could not resolve the default team for bare id "${n}" (${msg}) — pass an explicit id like LUM-${n}`);
+    }
+    const def = teams.find(t => t.isDefault) ?? (teams.length === 1 ? teams[0] : undefined);
+    if (!def) {
+        throw new Error(`workspace has multiple teams and no default — pass an explicit id like SPEC-${n} instead of a bare number`);
+    }
+    return `${def.identifier.toUpperCase()}-${n}`;
+}

package/dist/cli/src/lib/worktree.js

@@ -46,19 +46,21 @@ exports.getGitCommonDir = getGitCommonDir;
 exports.isMainCheckout = isMainCheckout;
 exports.getRepoRoot = getRepoRoot;
 /**
- * Normalize a user-supplied task reference to canonical `LUM-<n>` form.
- * Accepts `LUM-267`, `lum-267`, and bare `267`. Returns null for anything
- * that is not a recognizable task id.
+ * Normalize a user-supplied task reference to canonical `<TEAM>-<n>` form.
+ * Accepts any team prefix (`LUM-267`, `SPEC-12`, `spec-12`), not just LUM —
+ * the prefix is a team's configured identifier (1–10 letters), matching the
+ * single-sourced shape in shared/src/task-identifier.ts (LUM-419). The prefix
+ * is upper-cased; the number is preserved.
+ *
+ * A bare number (`267`) returns null here: without team context the prefix
+ * cannot be inferred locally. Team-aware bare-number resolution (via the
+ * workspace's default team) lives in `resolveWorktreeTaskId` (worktree-ref.ts).
  */
 function normalizeTaskId(input) {
-    const trimmed = input.trim();
-    const prefixed = trimmed.match(/^lum-?(\d+)$/i);
-    if (prefixed)
-        return `LUM-${prefixed[1]}`;
-    const bare = trimmed.match(/^(\d+)$/);
-    if (bare)
-        return `LUM-${bare[1]}`;
-    return null;
+    const m = input.trim().match(/^([A-Za-z]{1,10})-(\d+)$/);
+    if (!m)
+        return null;
+    return `${m[1].toUpperCase()}-${m[2]}`;
 }
 /**
  * Turn an arbitrary human slug into a branch/dir-safe segment: lowercase,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lumoai/cli",
-  "version": "1.56.0",
+  "version": "1.57.0",
   "description": "Lumo CLI — manage tasks and sessions from the terminal",
   "license": "MIT",
   "author": "cli@uselumo.ai",