@voltade/volty 0.1.9 → 0.1.10

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
@@ -46,7 +46,7 @@ Open Claude Code anywhere and ask in plain language. The skill you installed in
46
46
  - *"Why didn't the agent learn anything from that conversation?"*
47
47
  - *"Clean up the test conversations on prod."*
48
48
 
49
- Claude knows the safety rules: tests run against a sandboxed draft version of the agent with a synthetic contact, so nothing can ever be sent to a real contact, and Claude cleans up after itself when it's done.
49
+ Claude knows the safety rules: tests run on your workspace's Sandbox from a synthetic contact, whose replies are recorded in the thread but never sent, so nothing can reach a real contact. Claude cleans up after itself when it's done, and anything left over is removed after 24 hours.
50
50
 
51
51
  ## Upgrading
52
52
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@voltade/volty",
3
- "version": "0.1.9",
3
+ "version": "0.1.10",
4
4
  "description": "Remote CLI for the Volty platform — run operations, trigger real agent wakes, and inspect LLM I/O over an API key.",
5
5
  "license": "UNLICENSED",
6
6
  "type": "module",
@@ -23,6 +23,6 @@ Use the lane the behaviour was reported on. On the contact lane Volty declines t
23
23
  ## On prod
24
24
 
25
25
  - Name the org on every command (`--env prod --org <id>`, or `VOLTY_ENV` and `VOLTY_ORG` in the process), and read the stderr line the CLI prints before each command to check it names the org you meant. Act only in the org the user named.
26
- - Test through the sandbox, where every reply stays a draft. A real conversation is never a test target.
26
+ - Test through the Sandbox, where a reply to the synthetic contact is recorded as sent and never leaves. A real conversation is never a test target.
27
27
  - Run `volty debug sandbox-cleanup --all --env prod --org <id>` when you finish.
28
28
  - A `{pending:true}` result means the turn outran the wait, not that it failed: poll for it (`volty docs` names the command) instead of sending the message again.