@akshar5/cohall 0.4.9 → 0.5.0

package/CHANGELOG.md

@@ -1,5 +1,28 @@
 # Changelog
 
+## [0.5.0](https://github.com/AksharP5/cohall/compare/v0.4.10...v0.5.0) (2026-08-10)
+
+
+### Features
+
+* add safe all-device operations ([#47](https://github.com/AksharP5/cohall/issues/47)) ([34da34f](https://github.com/AksharP5/cohall/commit/34da34fd57cbde624f6c9f2dc171a2644a84faf0))
+* **cli:** make device setup guided ([#44](https://github.com/AksharP5/cohall/issues/44)) ([0330094](https://github.com/AksharP5/cohall/commit/033009416a97d6158af978589de92f77b4d1c3b7))
+* **relay:** make relay moves recoverable ([#46](https://github.com/AksharP5/cohall/issues/46)) ([851ddff](https://github.com/AksharP5/cohall/commit/851ddff4dcd8545477b729c8f5509a3b145fd759))
+
+
+### Bug Fixes
+
+* **device:** trust service manager executables ([#50](https://github.com/AksharP5/cohall/issues/50)) ([f150574](https://github.com/AksharP5/cohall/commit/f1505743621647e3ff8f239f2ebd6b11c9a9c09a))
+* **relay:** harden relay moves ([#48](https://github.com/AksharP5/cohall/issues/48)) ([fd8c8f3](https://github.com/AksharP5/cohall/commit/fd8c8f3f5d4a14deace8c82e3bcb043f33968127))
+* **relay:** make all-device maintenance recoverable ([#49](https://github.com/AksharP5/cohall/issues/49)) ([63fd304](https://github.com/AksharP5/cohall/commit/63fd304fe6c81c4d71122afe2bc2ca8899b0e117))
+
+## [0.4.10](https://github.com/AksharP5/cohall/compare/v0.4.9...v0.4.10) (2026-08-09)
+
+
+### Bug Fixes
+
+* **device:** preserve macOS provider startup ([#42](https://github.com/AksharP5/cohall/issues/42)) ([aef9f0d](https://github.com/AksharP5/cohall/commit/aef9f0db5ff2360e6bf38774d6b3350804b46f0e))
+
 ## [0.4.9](https://github.com/AksharP5/cohall/compare/v0.4.8...v0.4.9) (2026-08-09)
 
 

package/README.md

@@ -106,28 +106,71 @@ COHALL_TOKEN="$COHALL_TOKEN" \
 npx -y @akshar5/cohall pair --label "MacBook"
 ```
 
-Transfer the token privately. On the device being added:
+Transfer the token privately. On the device being added, one guided command
+configures the device and installs the agent skill. It defaults the device name
+to the hostname and prompts for omitted values when run interactively:
 
 ```bash
 read -rsp 'Pairing token: ' pairing_token; printf '\n'
-printf '%s' "$pairing_token" | npx -y @akshar5/cohall join \
+printf '%s' "$pairing_token" | npx -y @akshar5/cohall init \
   --relay https://cohall.example.com \
-  --name macbook \
   --providers codex \
   --workspace "$HOME/dev"
 unset pairing_token
 ```
 
-### 3. Install the skill and connect
+`cohall join` remains available for scripts that only want to exchange a
+pairing token and write configuration.
+
+### 3. Keep the device available
+
+```bash
+npm install --global --prefix "$HOME/.local" @akshar5/cohall
+cohall service install
+cohall doctor
+```
+
+The service installer uses the current user's systemd manager on Linux,
+launchd on macOS, and Task Scheduler on Windows. See
+[service setup](docs/services.md) for boot-before-login and relay-host details.
+
+## Move a relay
+
+Move an existing relay to an always-on machine without pairing everything
+again. On the current relay host:
 
 ```bash
-npx -y @akshar5/cohall skill install all
-npx -y @akshar5/cohall doctor
-npx -y @akshar5/cohall device
+cohall relay backup ./cohall-relay-backup
 ```
 
-Use an operating-system service for an unattended relay or device worker. See
-[service setup](docs/services.md).
+Copy that directory privately to the new host, install Cohall there, and restore
+it as the user and data directory that will run the relay:
+
+```bash
+cohall relay restore ./cohall-relay-backup
+cohall relay
+```
+
+After the new address is reachable, run this once on each configured client or
+device:
+
+```bash
+cohall relay use https://new-relay.example.com
+```
+
+The switch verifies the existing credentials against the restored relay before
+changing local configuration, then restarts an active device service. Remote
+addresses require HTTPS by default. Use `--allow-http` only when another layer,
+such as Tailscale, already encrypts the connection.
+
+The backup contains the SQLite relay state, the plaintext owner token, and a
+checksum manifest. Relay state includes devices, threads, prompts, results,
+provider-session references, upgrade history, pairing records, and hashed
+client/device sessions. It does not contain device-local files, provider
+credentials, Cohall device configuration, service definitions, or reverse-proxy
+TLS configuration. Transfer and store it like a secret. If the new host keeps
+the same DNS or Tailscale address, clients need no configuration change. See
+[moving a relay](docs/services.md#move-a-relay) for service-host details.
 
 ## Delegate work
 
@@ -150,6 +193,39 @@ npx -y @akshar5/cohall trace <task-id> --follow
 Reuse the returned `thread_id` for follow-ups so the target resumes its provider
 session.
 
+## Manage all devices
+
+Inspect the whole installation from any paired client:
+
+```bash
+cohall doctor --all
+cohall versions
+cohall usage
+```
+
+`doctor --all` reports connectivity, providers, workspaces, and version drift.
+`usage` reports retained Cohall task activity by device, status, and provider;
+provider token counts and billing are not available to the relay.
+
+From the relay owner account, queue a Cohall upgrade for every registered
+device and inspect progress:
+
+```bash
+cohall upgrade --all
+cohall upgrades
+# If a lost device can never finish its queued operation:
+cohall upgrades abandon <operation-id>
+```
+
+Upgrade requests are durable for offline devices and run after active delegated
+work. This is a typed owner-only maintenance operation, not remote shell access.
+Each target must use a global npm, Bun, or pnpm installation that can upgrade
+itself. Preview the targets with `cohall upgrade --all --dry-run`, or pin an
+exact release with `--to 1.2.3`. `cohall upgrades` returns the 50 newest results.
+Abandonment is an owner recovery action for a permanently unreachable target;
+it does not interrupt an upgrade that is already executing. Forgetting an
+offline device also closes its outstanding maintenance operation.
+
 ## Optional MCP
 
 CLI plus skill is the recommended integration. For a client that accepts the