packages/cli/README.md still lists `os studio` — the CLI ships no such command

[os-zhuang]

Filed by the triage seat as the upstream companion of objectstack-ai/objectos#150 (dedup-checked against open cards: os studio returns no open framework card; #10967 covers os projects residue in environments/*.ts, a different family).

Measured (by the objectos card, on origin/main @ 2866d5f97e)

• packages/cli/README.md:45 lists | os studio [config] | Launch Studio UI with development server |.
• The oclif command set is pattern-derived from packages/cli/src/commands/** and contains no studio command; a case-insensitive grep across packages/cli/src finds studio only in comments, i18n baselines and tests (control probe: the same grep does return dev.ts/serve.ts).
• serve.ts:2470-2472 states the current posture: @objectstack/studio is intentionally not default-loaded; the console ships /_console/studio/<pkg>/<pillar>.

Fix

Delete the README row (or replace with the accurate route: os dev / os serve, then /_console/studio in the browser). One-line docs fix, skip-changeset shape.

Refs: objectstack-ai/objectos#150 (downstream twin, queued there) · #10967 (adjacent stale-command-prose family).

[os-zhuang]

Claim: PM loop round 31 — family fold member; chain head is #11221

• Session: session_019siH5jDmk5hrayvfyojUqR (domain:cli lane execution seat, post [PM seat] domain:cli — 🟢 os-zhuang (session_019siH5jDmk5hrayvfyojUqR) · R32 · 落地 15 卡 / 13 PR · 在飞 4 · 待复审 1 · 待人工合并 1 · 决策箱 2 #6024)
• Branch: claude/issue-11221-cli-documented-commands-resolve — shared with the chain head. ⛔ This card has no branch of its own by design, not as a half-state: a folded member's independent-checkability lives in its own commit and its own Fixes line, not in a separate branch.
• Worktree: objectstack-issue-11221
• Domain: domain:cli
• File surface: packages/cli/README.md
• Container & model: folded into the chain head's M, mode:subagent, model: opus
• Clause-②: no — removing a README row for a command that does not exist. Nothing widens; nothing's accept/reject behaviour moves.
• Serial constraints: full check recorded on the chain head (packages/cli: register.ts/whoami.ts/logout.ts examples say os auth ... — that topic does not resolve (Command auth:whoami not found.) #11221 comment 5385065435) — ⛔ do not re-clear them here.

Fold rationale and the five gates are on the chain head. This card's own independent verdict: os studio gone from packages/cli/README.md, and confirmed absent from the built command registry (⛔ do not delete the row on the card's word alone — if os studio turns out to exist under another spelling, that is a different fix and a stop-and-report).

Measured at origin/main at dispatch: packages/cli/README.md:45 — | `os studio [config]` | Launch Studio UI with development server |. ⚠️ Re-derive the line number at your own head; #11266 merged into this area 8 minutes before this dispatch.

---

Generated by Claude Code

[os-zhuang]

✅ ACCEPT — PR #11314, folded member; full review on the chain head #11221 (5385449935)

domain:cli seat, session session_019siH5jDmk5hrayvfyojUqR. This card's own verdict, verified independently of the fold's:

item	verdict
The edit	packages/cli/README.md — exactly one line deleted, the os studio [config] row. Zero collateral edits to the table or anywhere else in the file ✅
Independent verifiability	yes — its own commit (a2f3c0bd), its own Fixes #11180 line, its own changeset ✅
Absence proof	two readings, not a grep: the built CLI's own oclif Config enumerates 60 ids with zero matching studio; and os studio --help exits 2 with Error: Command studio not found. ✅
Anti-vacuity	the enumeration carries a control — dev, serve, login, logout, register, whoami all present in the same list. That is what makes the zero evidence rather than a broken probe ✅

⭐ Deleted rather than rewritten, and the reasoning is right. Studio is not reached by a CLI command at all — the console serves it at /_console/studio after os dev or os serve, both already listed in that same table. A replacement row would have reintroduced the exact category error that made the original wrong: a Commands table lists commands, and a browser route is not one. Substituting a plausible-looking row is the easy wrong answer here, and it was declined with a stated reason.

⚠️ My dispatch warned that #11266 had probably moved line 45 and told the dev to re-derive. It had not — still exactly packages/cli/README.md:45. The dev checked and said so rather than silently accepting my framing. Recording it as a falsified PM assumption, not a defect.

Landing follows the chain head: ⛔ not enqueued until the named gate jobs reach completed: success on bc08f9bb.

---

Generated by Claude Code