@lmctl-ai/lmctl 0.1.82 → 0.1.84

package/CHANGELOG.md

@@ -2,6 +2,47 @@
 
 All notable public-preview changes for `@lmctl-ai/lmctl` are recorded here.
 
+## 0.1.84 — 2026-07-07
+
+- **lmplayer: `tail`/`health`/`ls` now find sessions on a fresh install.**
+  lmplayer writes a channel-suffixed session DB (`opencode-<channel>.db`, e.g.
+  `opencode-dev.db`) rather than `opencode-local.db`; the reader now scans all
+  lmplayer/opencode-compatible DB candidates and reads from the one that actually
+  contains the session (skipping stale/incompatible stores).
+- **Fixed the `model_selection.md not found` warning on global installs.** The
+  tested-models catalog is now shipped in the package, and the runtime resolver
+  finds it across source, package-root, and bundled-`dist` layouts; a
+  `provider=lmplayer model=…` member no longer emits a spurious "cannot be
+  verified" warning.
+
+## 0.1.83 — 2026-07-06
+
+- **`lmctl terminal` takes a DB lock** (host + pid + heartbeat) so a second chat
+  to a member held live in a terminal is rejected instead of racing; a stale/dead
+  or remote lock is reclaimed safely, and `--force` overrides.
+- **`--detach` is hidden from the CLI surface.** It stays functional for AI-team
+  chat but is the codex/gemini-only fallback; other providers use their native
+  tracked-background path.
+- **MCP install retired.** lmctl no longer writes provider MCP config on
+  seed/chat (the internal `lmctl_chat` code remains; only the installation is
+  removed).
+- **Skills over the network.** The default teamfile template pulls shared skills
+  (background-wakeup, team-lead-workflow, durable-memory) from lmctl.com via
+  `_SKILL_ <url>` into `durable-memory/skills/` on seed.
+- **`effort=` advisory for `provider=qwen`.** Qwen Code exposes no reasoning-effort
+  CLI flag, so lint points you to a thinking/effort model id instead of
+  mis-warning that effort is ignored.
+
+## 0.1.82 — 2026-07-06
+
+- **New first-class provider: `provider=lmplayer`.** A one-shot `lmplayer run`
+  adapter — seed captures the sessionID from the JSON stream, resume via
+  `run --session`, `effort=`, YOLO via `--dangerously-skip-permissions`, and it
+  reuses the opencode session-reader against lmplayer's store so
+  `tail`/`health`/`ls` work.
+- **`lmctl loop`** — a simple autopilot driver that repeatedly continues a member
+  until it reports `ALL DONE` or `OPERATOR ESCALATION`.
+
 ## Unreleased
 
 - Teamfile templates no longer hardcode `sessiondir`: `plan`/`hire` omit it

package/README.md

@@ -45,14 +45,14 @@ lmctl run https://lmctl.com/workflows/research.compound.json
 lmctl plan ./my-project
 lmctl lint ./my-project/my-project.lmctl
 lmctl seed ./my-project/my-project.lmctl
-lmctl chat ./my-project/my-project.lmctl Lead "Inspect the repo and propose next steps." --detach
-lmctl jobs watch <id>
+lmctl chat ./my-project/my-project.lmctl Lead "Inspect the repo and propose next steps."
 ```
 
 Start with `lmctl run <url>` for published workflows. For a custom AI-Lead team,
-use the plan -> lint -> seed -> `chat --detach` -> `jobs watch` flow so work is
-tracked and recoverable. Daemon, web, API, and MCP surfaces are secondary
-integration paths; the CLI/job portal is the golden path for onboarding.
+use the plan -> lint -> seed -> chat flow. Use provider-native background
+execution for long turns, with `lmctl jobs` as the tracked work portal. Daemon,
+web, API, and MCP surfaces are secondary integration paths; the CLI/job portal
+is the golden path for onboarding.
 Lintable teamfile templates for research, writing, and review teams ship in the
 [examples gallery](https://lmctl.com/examples/).