@lmctl-ai/lmctl 0.1.83 → 0.1.84
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +41 -0
- package/dist/cli/index.js +475 -475
- package/durable-memory/model_selection.md +138 -0
- package/package.json +2 -1
package/CHANGELOG.md
CHANGED
|
@@ -2,6 +2,47 @@
|
|
|
2
2
|
|
|
3
3
|
All notable public-preview changes for `@lmctl-ai/lmctl` are recorded here.
|
|
4
4
|
|
|
5
|
+
## 0.1.84 — 2026-07-07
|
|
6
|
+
|
|
7
|
+
- **lmplayer: `tail`/`health`/`ls` now find sessions on a fresh install.**
|
|
8
|
+
lmplayer writes a channel-suffixed session DB (`opencode-<channel>.db`, e.g.
|
|
9
|
+
`opencode-dev.db`) rather than `opencode-local.db`; the reader now scans all
|
|
10
|
+
lmplayer/opencode-compatible DB candidates and reads from the one that actually
|
|
11
|
+
contains the session (skipping stale/incompatible stores).
|
|
12
|
+
- **Fixed the `model_selection.md not found` warning on global installs.** The
|
|
13
|
+
tested-models catalog is now shipped in the package, and the runtime resolver
|
|
14
|
+
finds it across source, package-root, and bundled-`dist` layouts; a
|
|
15
|
+
`provider=lmplayer model=…` member no longer emits a spurious "cannot be
|
|
16
|
+
verified" warning.
|
|
17
|
+
|
|
18
|
+
## 0.1.83 — 2026-07-06
|
|
19
|
+
|
|
20
|
+
- **`lmctl terminal` takes a DB lock** (host + pid + heartbeat) so a second chat
|
|
21
|
+
to a member held live in a terminal is rejected instead of racing; a stale/dead
|
|
22
|
+
or remote lock is reclaimed safely, and `--force` overrides.
|
|
23
|
+
- **`--detach` is hidden from the CLI surface.** It stays functional for AI-team
|
|
24
|
+
chat but is the codex/gemini-only fallback; other providers use their native
|
|
25
|
+
tracked-background path.
|
|
26
|
+
- **MCP install retired.** lmctl no longer writes provider MCP config on
|
|
27
|
+
seed/chat (the internal `lmctl_chat` code remains; only the installation is
|
|
28
|
+
removed).
|
|
29
|
+
- **Skills over the network.** The default teamfile template pulls shared skills
|
|
30
|
+
(background-wakeup, team-lead-workflow, durable-memory) from lmctl.com via
|
|
31
|
+
`_SKILL_ <url>` into `durable-memory/skills/` on seed.
|
|
32
|
+
- **`effort=` advisory for `provider=qwen`.** Qwen Code exposes no reasoning-effort
|
|
33
|
+
CLI flag, so lint points you to a thinking/effort model id instead of
|
|
34
|
+
mis-warning that effort is ignored.
|
|
35
|
+
|
|
36
|
+
## 0.1.82 — 2026-07-06
|
|
37
|
+
|
|
38
|
+
- **New first-class provider: `provider=lmplayer`.** A one-shot `lmplayer run`
|
|
39
|
+
adapter — seed captures the sessionID from the JSON stream, resume via
|
|
40
|
+
`run --session`, `effort=`, YOLO via `--dangerously-skip-permissions`, and it
|
|
41
|
+
reuses the opencode session-reader against lmplayer's store so
|
|
42
|
+
`tail`/`health`/`ls` work.
|
|
43
|
+
- **`lmctl loop`** — a simple autopilot driver that repeatedly continues a member
|
|
44
|
+
until it reports `ALL DONE` or `OPERATOR ESCALATION`.
|
|
45
|
+
|
|
5
46
|
## Unreleased
|
|
6
47
|
|
|
7
48
|
- Teamfile templates no longer hardcode `sessiondir`: `plan`/`hire` omit it
|