@use-aistack/cli 0.10.1 → 0.11.0

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/README.md CHANGED
@@ -20,17 +20,19 @@ npx @use-aistack/cli sync
20
20
 
21
21
  On an unlinked machine, `sync` starts the login flow inline - your browser opens to approve the machine, then the sync continues. One command is the whole onboarding.
22
22
 
23
+ After a manual sync, the CLI offers auto-sync with three choices: Enable, Maybe later, and Never ask again. Maybe later asks again after your next manual sync. Existing declines from older CLI versions are treated as Maybe later.
24
+
23
25
  ### `npx @use-aistack/cli sync --auto on` / `off`
24
26
 
25
- Optional: keep your stack fresh without manual syncs. `on` asks your stack for the permission, then writes a `SessionStart` hook into the harnesses you actually use - `~/.claude/settings.json` for Claude Code, `~/.codex/hooks.json` for Codex. The hook runs a silent sync at most once per day when a session starts. `off` removes the hooks and takes the permission back.
27
+ Optional: keep your stack fresh without manual syncs. `on` asks your stack for the permission, then writes a `SessionStart` hook into the harnesses you actually use - `~/.claude/settings.json` for Claude Code, `~/.codex/hooks.json` for Codex. The hook runs a silent sync at most once every 6 hours when a session starts. `off` removes the hooks and takes the permission back.
26
28
 
27
29
  ```sh
28
- npx @use-aistack/cli sync --auto on # enable, default every 24h
30
+ npx @use-aistack/cli sync --auto on # enable, default every 6h
29
31
  npx @use-aistack/cli sync --auto on --every 12 # custom frequency in hours
30
32
  npx @use-aistack/cli sync --auto off # revoke
31
33
  ```
32
34
 
33
- **Your stack owns the permission, not this machine.** The silent run asks aistack.to before it publishes anything, so the switch on your stack page is a complete revoke: it stops every machine, even one whose hooks are still installed. It works the other way too - turn the switch on, run one `sync`, and that machine installs its own triggers. A harness you adopt months later gets its trigger the same way.
35
+ **Your stack owns the permission, not this machine.** The silent run asks aistack.to before it publishes anything, so the switch above the first section on your stack page is a complete revoke: it stops every machine, even one whose hooks are still installed. The switch defaults to every 6 hours. Run the enable command shown beside it on each machine that should install triggers. A harness you adopt months later gets its trigger the same way.
34
36
 
35
37
  The silent run (`sync --auto`) never prompts and never installs a hook. Each run appends one line to `~/.config/aistack/sync.log` (capped at 200 lines). The next interactive `sync` reports the last result. After 3 failures in a row, one visible message appears in Claude Code and names the fix. No email, no dialogs.
36
38
 
@@ -40,6 +42,7 @@ Link this machine to your AI Stack account via browser. Optional - `sync` runs t
40
42
 
41
43
  ```sh
42
44
  npx @use-aistack/cli login
45
+ npx @use-aistack/cli login --label "build server" # preset and lock the label
43
46
  ```
44
47
 
45
48
  ### `npx @use-aistack/cli connect claude`